Artificial Intelligence Nanodegree

Computer Vision Capstone

Project: Facial Keypoint Detection

---

Welcome to the final Computer Vision project in the Artificial Intelligence Nanodegree program!

In this project, you’ll combine your knowledge of computer vision techniques and deep learning to build and end-to-end facial keypoint recognition system! Facial keypoints include points around the eyes, nose, and mouth on any face and are used in many applications, from facial tracking to emotion recognition.

There are three main parts to this project:

Part 1 : Investigating OpenCV, pre-processing, and face detection

Part 2 : Training a Convolutional Neural Network (CNN) to detect facial keypoints

Part 3 : Putting parts 1 and 2 together to identify facial keypoints on any image!

---

*Here's what you need to know to complete the project:

1. In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested.

   a. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!

1. In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation.

   a. Each section where you will answer a question is preceded by a 'Question X' header.

   b. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'.

Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.

The rubric contains optional suggestions for enhancing the project beyond the minimum requirements. If you decide to pursue the "(Optional)" sections, you should include the code in this IPython notebook.

Your project submission will be evaluated based on your answers to each of the questions and the code implementations you provide.

Steps to Complete the Project

Each part of the notebook is further broken down into separate steps. Feel free to use the links below to navigate the notebook.

In this project you will get to explore a few of the many computer vision algorithms built into the OpenCV library. This expansive computer vision library is now almost 20 years old and still growing!

The project itself is broken down into three large parts, then even further into separate steps. Make sure to read through each step, and complete any sections that begin with '(IMPLEMENTATION)' in the header; these implementation sections may contain multiple TODOs that will be marked in code. For convenience, we provide links to each of these steps below.

Part 1 : Investigating OpenCV, pre-processing, and face detection

• Step 0: Detect Faces Using a Haar Cascade Classifier
• Step 1: Add Eye Detection
• Step 2: De-noise an Image for Better Face Detection
• Step 3: Blur an Image and Perform Edge Detection
• Step 4: Automatically Hide the Identity of an Individual

Part 2 : Training a Convolutional Neural Network (CNN) to detect facial keypoints

• Step 5: Create a CNN to Recognize Facial Keypoints
• Step 6: Compile and Train the Model
• Step 7: Visualize the Loss and Answer Questions

Part 3 : Putting parts 1 and 2 together to identify facial keypoints on any image!

• Step 8: Build a Robust Facial Keypoints Detector (Complete the CV Pipeline)

---

Step 0: Detect Faces Using a Haar Cascade Classifier

Have you ever wondered how Facebook automatically tags images with your friends' faces? Or how high-end cameras automatically find and focus on a certain person's face? Applications like these depend heavily on the machine learning task known as face detection - which is the task of automatically finding faces in images containing people.

At its root face detection is a classification problem - that is a problem of distinguishing between distinct classes of things. With face detection these distinct classes are 1) images of human faces and 2) everything else.

We use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images. OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the detector_architectures directory.

Import Resources

In the next python cell, we load in the required libraries for this section of the project.

# Import required libraries for this section

%matplotlib inline

import numpy as np
import matplotlib.pyplot as plt
import math
import cv2                     # OpenCV library for computer vision
from PIL import Image
import time 

Next, we load in and display a test image for performing face detection.

Note: by default OpenCV assumes the ordering of our image's color channels are Blue, then Green, then Red. This is slightly out of order with most image types we'll use in these experiments, whose color channels are ordered Red, then Green, then Blue. In order to switch the Blue and Red channels of our test image around we will use OpenCV's cvtColor function, which you can read more about by checking out some of its documentation located here. This is a general utility function that can do other transformations too like converting a color image to grayscale, and transforming a standard color image to HSV color space.

# Load in color image for face detection
image = cv2.imread('images/test_image_1.jpg')

# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Plot our image using subplots to specify a size and title
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Original Image')
ax1.imshow(image)

<matplotlib.image.AxesImage at 0x7f800eee9828>

There are a lot of people - and faces - in this picture. 13 faces to be exact! In the next code cell, we demonstrate how to use a Haar Cascade classifier to detect all the faces in this test image.

This face detector uses information about patterns of intensity in an image to reliably detect faces under varying light conditions. So, to use this face detector, we'll first convert the image from color to grayscale.

Then, we load in the fully trained architecture of the face detector -- found in the file haarcascade_frontalface_default.xml - and use it on our image to find faces!

To learn more about the parameters of the detector see this post.

# Convert the RGB  image to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)

# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

# Detect the faces in image
faces = face_cascade.detectMultiScale(gray, 4, 6)

# Print the number of faces detected in the image
print('Number of faces detected:', len(faces))

# Make a copy of the orginal image to draw face detections on
image_with_detections = np.copy(image)

# Get the bounding box for each detected face
for (x,y,w,h) in faces:
    # Add a red bounding box to the detections image
    cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
    

# Display the image with the detections
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Image with Face Detections')
ax1.imshow(image_with_detections)

Number of faces detected: 13

<matplotlib.image.AxesImage at 0x7f803bf80d30>

In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.

---

Step 1: Add Eye Detections

There are other pre-trained detectors available that use a Haar Cascade Classifier - including full human body detectors, license plate detectors, and more. A full list of the pre-trained architectures can be found here.

To test your eye detector, we'll first read in a new test image with just a single face.

# Load in color image for face detection
image = cv2.imread('images/james.jpg')

# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Plot the RGB image
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Original Image')
ax1.imshow(image)

<matplotlib.image.AxesImage at 0x7f803bf2f080>

Notice that even though the image is a black and white image, we have read it in as a color image and so it will still need to be converted to grayscale in order to perform the most accurate face detection.

So, the next steps will be to convert this image to grayscale, then load OpenCV's face detector and run it with parameters that detect this face accurately.

# Convert the RGB  image to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)

# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

# Detect the faces in image
faces = face_cascade.detectMultiScale(gray, 1.25, 6)

# Print the number of faces detected in the image
print('Number of faces detected:', len(faces))

# Make a copy of the orginal image to draw face detections on
image_with_detections = np.copy(image)

# Get the bounding box for each detected face
for (x,y,w,h) in faces:
    # Add a red bounding box to the detections image
    cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
    

# Display the image with the detections
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Image with Face Detection')
ax1.imshow(image_with_detections)

Number of faces detected: 1

<matplotlib.image.AxesImage at 0x7f803bf5a048>

(IMPLEMENTATION) Add an eye detector to the current face detection setup.

A Haar-cascade eye detector can be included in the same way that the face detector was and, in this first task, it will be your job to do just this.

To set up an eye detector, use the stored parameters of the eye cascade detector, called haarcascade_eye.xml, located in the detector_architectures subdirectory. In the next code cell, create your eye detector and store its detections.

A few notes before you get started:

First, make sure to give your loaded eye detector the variable name

eye_cascade

and give the list of eye regions you detect the variable name

eyes

Second, since we've already run the face detector over this image, you should only search for eyes within the rectangular face regions detected in faces. This will minimize false detections.

Lastly, once you've run your eye detector over the facial detection region, you should display the RGB image with both the face detection boxes (in red) and your eye detections (in green) to verify that everything works as expected.

# Make a copy of the original image to plot rectangle detections
image_with_detections = np.copy(image)   

# Loop over the detections and draw their corresponding face detection boxes
for (x,y,w,h) in faces:
    cv2.rectangle(image_with_detections, (x,y), (x+w,y+h),(255,0,0), 3)  
    
# Do not change the code above this comment!

    
## TODO: Add eye detection, using haarcascade_eye.xml, to the current face detector algorithm
eye_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_eye.xml')

## TODO: Loop over the eye detections and draw their corresponding boxes in green on image_with_detections
for (x_f,y_f,w_f,h_f) in faces:
    # detect eyes in each detected face
    eyes = eye_cascade.detectMultiScale(image_with_detections[y_f:y_f+h_f,x_f:x_f+w_f], 1.2, 6)
    # draw green rectangle for each eyes
    for (x_e,y_e,w_e,h_e) in eyes:
        # absolute position of eyes
        x_e_abs = x_f+x_e
        y_e_abs = y_f+y_e
        # draw the rectangle
        cv2.rectangle(image_with_detections,(x_e_abs,y_e_abs), (x_e_abs+w_e, y_e_abs+h_e), (0,255,0), 3)
        

# Plot the image with both faces and eyes detected
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Image with Face and Eye Detection')
ax1.imshow(image_with_detections)

<matplotlib.image.AxesImage at 0x7f803bf06550>

(Optional) Add face and eye detection to your laptop camera

It's time to kick it up a notch, and add face and eye detection to your laptop's camera! Afterwards, you'll be able to show off your creation like in the gif shown below - made with a completed version of the code!

Notice that not all of the detections here are perfect - and your result need not be perfect either. You should spend a small amount of time tuning the parameters of your detectors to get reasonable results, but don't hold out for perfection. If we wanted perfection we'd need to spend a ton of time tuning the parameters of each detector, cleaning up the input image frames, etc. You can think of this as more of a rapid prototype.

The next cell contains code for a wrapper function called laptop_camera_face_eye_detector that, when called, will activate your laptop's camera. You will place the relevant face and eye detection code in this wrapper function to implement face/eye detection and mark those detections on each image frame that your camera captures.

Before adding anything to the function, you can run it to get an idea of how it works - a small window should pop up showing you the live feed from your camera; you can press any key to close this window.

Note: Mac users may find that activating this function kills the kernel of their notebook every once in a while. If this happens to you, just restart your notebook's kernel, activate cell(s) containing any crucial import statements, and you'll be good to go!

### Add face and eye detection to this laptop camera function 
# Make sure to draw out all faces/eyes found in each frame on the shown video feed

import cv2
import time 

# wrapper function for face/eye detection with your laptop camera
def laptop_camera_go():
    # Create instance of video capturer
    cv2.namedWindow("face detection activated")
    vc = cv2.VideoCapture(0)

    # Try to get the first frame
    if vc.isOpened(): 
        rval, frame = vc.read()
    else:
        rval = False
    
    # Keep the video stream open
    while rval:
        # Plot the image from camera with all the face and eye detections marked
        cv2.imshow("face detection activated", frame)
        
        # Exit functionality - press any key to exit laptop video
        key = cv2.waitKey(20)
        if key > 0: # Exit by pressing any key
            # Destroy windows 
            cv2.destroyAllWindows()
            
            # Make sure window closes on OSx
            for i in range (1,5):
                cv2.waitKey(1)
            return
        
        # Read next frame
        time.sleep(0.05)             # control framerate for computation - default 20 frames per sec
        rval, frame = vc.read()    

# Call the laptop camera face/eye detector function above
laptop_camera_go()

---

Step 2: De-noise an Image for Better Face Detection

Image quality is an important aspect of any computer vision task. Typically, when creating a set of images to train a deep learning network, significant care is taken to ensure that training images are free of visual noise or artifacts that hinder object detection. While computer vision algorithms - like a face detector - are typically trained on 'nice' data such as this, new test data doesn't always look so nice!

When applying a trained computer vision algorithm to a new piece of test data one often cleans it up first before feeding it in. This sort of cleaning - referred to as pre-processing - can include a number of cleaning phases like blurring, de-noising, color transformations, etc., and many of these tasks can be accomplished using OpenCV.

In this short subsection we explore OpenCV's noise-removal functionality to see how we can clean up a noisy image, which we then feed into our trained face detector.

Create a noisy image to work with

In the next cell, we create an artificial noisy version of the previous multi-face image. This is a little exaggerated - we don't typically get images that are this noisy - but image noise, or 'grainy-ness' in a digitial image - is a fairly common phenomenon.

# Load in the multi-face test image again
image = cv2.imread('images/test_image_1.jpg')

# Convert the image copy to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Make an array copy of this image
image_with_noise = np.asarray(image)

# Create noise - here we add noise sampled randomly from a Gaussian distribution: a common model for noise
noise_level = 40
noise = np.random.randn(image.shape[0],image.shape[1],image.shape[2])*noise_level

# Add this noise to the array image copy
image_with_noise = image_with_noise + noise

# Convert back to uint8 format
image_with_noise = np.asarray([np.uint8(np.clip(i,0,255)) for i in image_with_noise])

# Plot our noisy image!
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Noisy Image')
ax1.imshow(image_with_noise)

<matplotlib.image.AxesImage at 0x7f7ff62b68d0>

In the context of face detection, the problem with an image like this is that - due to noise - we may miss some faces or get false detections.

In the next cell we apply the same trained OpenCV detector with the same settings as before, to see what sort of detections we get.

# Convert the RGB  image to grayscale
gray_noise = cv2.cvtColor(image_with_noise, cv2.COLOR_RGB2GRAY)

# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

# Detect the faces in image
faces = face_cascade.detectMultiScale(gray_noise, 4, 6)

# Print the number of faces detected in the image
print('Number of faces detected:', len(faces))

# Make a copy of the orginal image to draw face detections on
image_with_detections = np.copy(image_with_noise)

# Get the bounding box for each detected face
for (x,y,w,h) in faces:
    # Add a red bounding box to the detections image
    cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
    

# Display the image with the detections
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Noisy Image with Face Detections')
ax1.imshow(image_with_detections)

Number of faces detected: 12

<matplotlib.image.AxesImage at 0x7f7ff62620b8>

With this added noise we now miss one of the faces!

(IMPLEMENTATION) De-noise this image for better face detection

Time to get your hands dirty: using OpenCV's built in color image de-noising functionality called fastNlMeansDenoisingColored - de-noise this image enough so that all the faces in the image are properly detected. Once you have cleaned the image in the next cell, use the cell that follows to run our trained face detector over the cleaned image to check out its detections.

You can find its official documentation here and a useful example here.

Note: you can keep all parameters except photo_render fixed as shown in the second link above. Play around with the value of this parameter - see how it affects the resulting cleaned image.

## TODO: Use OpenCV's built in color image de-noising function to clean up our noisy image!

def denoising_image(image):
    return cv2.fastNlMeansDenoisingColored(image,None,18,21,7,21)

denoised_image = denoising_image(image_with_noise)

# Display the denoised image
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('DeNoised Image')
ax1.imshow(denoised_image)

<matplotlib.image.AxesImage at 0x7f7ff627feb8>

## TODO: Run the face detector on the de-noised image to improve your detections and display the result

# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

# Detect the faces in image
faces = face_cascade.detectMultiScale(denoised_image, 4, 6)

# Print the number of faces detected in the image
print('Number of faces detected:', len(faces))

# Make a copy of the orginal image to draw face detections on
denoised_image_with_detections = np.copy(denoised_image)

# Get the bounding box for each detected face
for (x,y,w,h) in faces:
    # Add a red bounding box to the detections image
    cv2.rectangle(denoised_image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
    

# Display the image with the detections
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('DeNoised Image with Face Detections')
ax1.imshow(denoised_image_with_detections)

Number of faces detected: 13

<matplotlib.image.AxesImage at 0x7f7ff62365c0>

---

Step 3: Blur an Image and Perform Edge Detection

Now that we have developed a simple pipeline for detecting faces using OpenCV - let's start playing around with a few fun things we can do with all those detected faces!

Importance of Blur in Edge Detection

Edge detection is a concept that pops up almost everywhere in computer vision applications, as edge-based features (as well as features built on top of edges) are often some of the best features for e.g., object detection and recognition problems.

Edge detection is a dimension reduction technique - by keeping only the edges of an image we get to throw away a lot of non-discriminating information. And typically the most useful kind of edge-detection is one that preserves only the important, global structures (ignoring local structures that aren't very discriminative). So removing local structures / retaining global structures is a crucial pre-processing step to performing edge detection in an image, and blurring can do just that.

Below is an animated gif showing the result of an edge-detected cat taken from Wikipedia, where the image is gradually blurred more and more prior to edge detection. When the animation begins you can't quite make out what it's a picture of, but as the animation evolves and local structures are removed via blurring the cat becomes visible in the edge-detected image.

Edge detection is a convolution performed on the image itself, and you can read about Canny edge detection on this OpenCV documentation page.

Canny edge detection

In the cell below we load in a test image, then apply Canny edge detection on it. The original image is shown on the left panel of the figure, while the edge-detected version of the image is shown on the right. Notice how the result looks very busy - there are too many little details preserved in the image before it is sent to the edge detector. When applied in computer vision applications, edge detection should preserve global structure; doing away with local structures that don't help describe what objects are in the image.

# Load in the image
image = cv2.imread('images/fawzia.jpg')

# Convert to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Convert to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)  

# Perform Canny edge detection
edges = cv2.Canny(gray,100,200)

# Dilate the image to amplify edges
edges = cv2.dilate(edges, None)

# Plot the RGB and edge-detected image
fig = plt.figure(figsize = (15,15))
ax1 = fig.add_subplot(121)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Original Image')
ax1.imshow(image)

ax2 = fig.add_subplot(122)
ax2.set_xticks([])
ax2.set_yticks([])

ax2.set_title('Canny Edges')
ax2.imshow(edges, cmap='gray')

<matplotlib.image.AxesImage at 0x7f7ff6207160>

Without first blurring the image, and removing small, local structures, a lot of irrelevant edge content gets picked up and amplified by the detector (as shown in the right panel above).

(IMPLEMENTATION) Blur the image then perform edge detection

In the next cell, you will repeat this experiment - blurring the image first to remove these local structures, so that only the important boudnary details remain in the edge-detected image.

Blur the image by using OpenCV's filter2d functionality - which is discussed in this documentation page - and use an averaging kernel of width equal to 4.

### TODO: Blur the test imageusing OpenCV's filter2d functionality, 
# Use an averaging kernel, and a kernel width equal to 4
def blur_image(image, kernel_size):
    kernel = np.ones((kernel_size,kernel_size), np.float32)/(kernel_size*kernel_size)
    return cv2.filter2D(image, -1, kernel)

blur = blur_image(image, 4)

## TODO: Then perform Canny edge detection and display the output
# Convert to RGB colorspace
gray = cv2.cvtColor(blur, cv2.COLOR_RGB2GRAY)  

# Perform Canny edge detection
edges = cv2.Canny(gray,100,200)

# Dilate the image to amplify edges
edges = cv2.dilate(edges, None)

# Plot the RGB and edge-detected image
fig = plt.figure(figsize = (15,15))
ax1 = fig.add_subplot(121)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Original Image')
ax1.imshow(image)

ax2 = fig.add_subplot(122)
ax2.set_xticks([])
ax2.set_yticks([])

ax2.set_title('Blurred Canny Edges')
ax2.imshow(edges, cmap='gray')

<matplotlib.image.AxesImage at 0x7f7ff450a2e8>

---

Step 4: Automatically Hide the Identity of an Individual

If you film something like a documentary or reality TV, you must get permission from every individual shown on film before you can show their face, otherwise you need to blur it out - by blurring the face a lot (so much so that even the global structures are obscured)! This is also true for projects like Google's StreetView maps - an enormous collection of mapping images taken from a fleet of Google vehicles. Because it would be impossible for Google to get the permission of every single person accidentally captured in one of these images they blur out everyone's faces, the detected images must automatically blur the identity of detected people. Here's a few examples of folks caught in the camera of a Google street view vehicle.

Read in an image to perform identity detection

Let's try this out for ourselves. Use the face detection pipeline built above and what you know about using the filter2D to blur and image, and use these in tandem to hide the identity of the person in the following image - loaded in and printed in the next cell.

# Load in the image
image = cv2.imread('images/gus.jpg')

# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Display the image
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Original Image')
ax1.imshow(image)

<matplotlib.image.AxesImage at 0x7f7ff45313c8>

(IMPLEMENTATION) Use blurring to hide the identity of an individual in an image

The idea here is to 1) automatically detect the face in this image, and then 2) blur it out! Make sure to adjust the parameters of the averaging blur filter to completely obscure this person's identity.

## TODO: Implement face detection

# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

blurred_image = np.copy(image)

# denoising image
denoised_image = denoising_image(blurred_image)

# Convert the RGB  image to grayscale
gray = cv2.cvtColor(denoised_image, cv2.COLOR_RGB2GRAY)

# Detect the faces in image
faces = face_cascade.detectMultiScale(gray, 4, 6)

## TODO: Blur the bounding box around each detected face using an averaging filter and display the result
for (x,y,w,h) in faces:
    face = blurred_image[y:y+h,x:x+w]
    blurred_image[y:y+h,x:x+w] = blur_image(face,95)

# Display the image
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Blurred Faces Image')
ax1.imshow(blurred_image)

<matplotlib.image.AxesImage at 0x7f7ff44dab00>

(Optional) Build identity protection into your laptop camera

In this optional task you can add identity protection to your laptop camera, using the previously completed code where you added face detection to your laptop camera - and the task above. You should be able to get reasonable results with little parameter tuning - like the one shown in the gif below.

As with the previous video task, to make this perfect would require significant effort - so don't strive for perfection here, strive for reasonable quality.

The next cell contains code a wrapper function called laptop_camera_identity_hider that - when called - will activate your laptop's camera. You need to place the relevant face detection and blurring code developed above in this function in order to blur faces entering your laptop camera's field of view.

Before adding anything to the function you can call it to get a hang of how it works - a small window will pop up showing you the live feed from your camera, you can press any key to close this window.

Note: Mac users may find that activating this function kills the kernel of their notebook every once in a while. If this happens to you, just restart your notebook's kernel, activate cell(s) containing any crucial import statements, and you'll be good to go!

### Insert face detection and blurring code into the wrapper below to create an identity protector on your laptop!
import cv2
import time 

def laptop_camera_go():
    # Create instance of video capturer
    cv2.namedWindow("face detection activated")
    vc = cv2.VideoCapture(0)

    # Try to get the first frame
    if vc.isOpened(): 
        rval, frame = vc.read()
    else:
        rval = False
    
    # Keep video stream open
    while rval:
        # Plot image from camera with detections marked
        cv2.imshow("face detection activated", frame)
        
        # Exit functionality - press any key to exit laptop video
        key = cv2.waitKey(20)
        if key > 0: # Exit by pressing any key
            # Destroy windows
            cv2.destroyAllWindows()
            
            for i in range (1,5):
                cv2.waitKey(1)
            return
        
        # Read next frame
        time.sleep(0.05)             # control framerate for computation - default 20 frames per sec
        rval, frame = vc.read()    
        

# Run laptop identity hider
laptop_camera_go()

---

Step 5: Create a CNN to Recognize Facial Keypoints

OpenCV is often used in practice with other machine learning and deep learning libraries to produce interesting results. In this stage of the project you will create your own end-to-end pipeline - employing convolutional networks in keras along with OpenCV - to apply a "selfie" filter to streaming video and images.

You will start by creating and then training a convolutional network that can detect facial keypoints in a small dataset of cropped images of human faces. We then guide you towards OpenCV to expanding your detection algorithm to more general images. What are facial keypoints? Let's take a look at some examples.

Facial keypoints (also called facial landmarks) are the small blue-green dots shown on each of the faces in the image above - there are 15 keypoints marked in each image. They mark important areas of the face - the eyes, corners of the mouth, the nose, etc. Facial keypoints can be used in a variety of machine learning applications from face and emotion recognition to commercial applications like the image filters popularized by Snapchat.

Below we illustrate a filter that, using the results of this section, automatically places sunglasses on people in images (using the facial keypoints to place the glasses correctly on each face). Here, the facial keypoints have been colored lime green for visualization purposes.

Make a facial keypoint detector

But first things first: how can we make a facial keypoint detector? Well, at a high level, notice that facial keypoint detection is a regression problem. A single face corresponds to a set of 15 facial keypoints (a set of 15 corresponding $(x, y)$ coordinates, i.e., an output point). Because our input data are images, we can employ a convolutional neural network to recognize patterns in our images and learn how to identify these keypoint given sets of labeled data.

In order to train a regressor, we need a training set - a set of facial image / facial keypoint pairs to train on. For this we will be using this dataset from Kaggle. We've already downloaded this data and placed it in the data directory. Make sure that you have both the training and test data files. The training dataset contains several thousand $96 \times 96$ grayscale images of cropped human faces, along with each face's 15 corresponding facial keypoints (also called landmarks) that have been placed by hand, and recorded in $(x, y)$ coordinates. This wonderful resource also has a substantial testing set, which we will use in tinkering with our convolutional network.

To load in this data, run the Python cell below - notice we will load in both the training and testing sets.

The load_data function is in the included utils.py file.

from utils import *

# Load training set
X_train, y_train = load_data()
print("X_train.shape == {}".format(X_train.shape))
print("y_train.shape == {}; y_train.min == {:.3f}; y_train.max == {:.3f}".format(
    y_train.shape, y_train.min(), y_train.max()))

# Load testing set
X_test, _ = load_data(test=True)
print("X_test.shape == {}".format(X_test.shape))

X_train.shape == (2140, 96, 96, 1)
y_train.shape == (2140, 30); y_train.min == -0.920; y_train.max == 0.996
X_test.shape == (1783, 96, 96, 1)

The load_data function in utils.py originates from this excellent blog post, which you are strongly encouraged to read. Please take the time now to review this function. Note how the output values - that is, the coordinates of each set of facial landmarks - have been normalized to take on values in the range $[-1, 1]$, while the pixel values of each input point (a facial image) have been normalized to the range $[0,1]$.

Note: the original Kaggle dataset contains some images with several missing keypoints. For simplicity, the load_data function removes those images with missing labels from the dataset. As an optional extension, you are welcome to amend the load_data function to include the incomplete data points.

Visualize the Training Data

Execute the code cell below to visualize a subset of the training data.

import matplotlib.pyplot as plt
%matplotlib inline

fig = plt.figure(figsize=(20,20))
fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)
for i in range(9):
    ax = fig.add_subplot(3, 3, i + 1, xticks=[], yticks=[])
    plot_data(X_train[i], y_train[i], ax)

For each training image, there are two landmarks per eyebrow (four total), three per eye (six total), four for the mouth, and one for the tip of the nose.

Review the plot_data function in utils.py to understand how the 30-dimensional training labels in y_train are mapped to facial locations, as this function will prove useful for your pipeline.

(IMPLEMENTATION) Specify the CNN Architecture

In this section, you will specify a neural network for predicting the locations of facial keypoints. Use the code cell below to specify the architecture of your neural network. We have imported some layers that you may find useful for this task, but if you need to use more Keras layers, feel free to import them in the cell.

Your network should accept a $96 \times 96$ grayscale image as input, and it should output a vector with 30 entries, corresponding to the predicted (horizontal and vertical) locations of 15 facial keypoints. If you are not sure where to start, you can find some useful starting architectures in this blog, but you are not permitted to copy any of the architectures that you find online.

# Import deep learning resources from Keras
from keras.models import Sequential
from keras.layers import Convolution2D, MaxPooling2D, Dropout
from keras.layers import Flatten, Dense


## TODO: Specify a CNN architecture
# Your model should accept 96x96 pixel graysale images in
# It should have a fully-connected output layer with 30 values (2 for each facial keypoint)

model = Sequential()
model.add(Convolution2D(filters=64, kernel_size=2, padding='same', activation='relu', input_shape=(96,96,1)))
model.add(MaxPooling2D(pool_size=2))
model.add(Dropout(0.3))

model.add(Convolution2D(filters=128, kernel_size=2, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))
model.add(Dropout(0.3))

model.add(Convolution2D(filters=256, kernel_size=2, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))
model.add(Dropout(0.3))

model.add(Flatten())
model.add(Dense(256, activation='relu'))
model.add(Dropout(0.4))
model.add(Dense(150, activation='relu'))
model.add(Dropout(0.4))
model.add(Dense(30, activation='tanh'))



# Summarize the model
model.summary()

Using TensorFlow backend.

_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv2d_1 (Conv2D)            (None, 96, 96, 64)        320       
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 48, 48, 64)        0         
_________________________________________________________________
dropout_1 (Dropout)          (None, 48, 48, 64)        0         
_________________________________________________________________
conv2d_2 (Conv2D)            (None, 48, 48, 128)       32896     
_________________________________________________________________
max_pooling2d_2 (MaxPooling2 (None, 24, 24, 128)       0         
_________________________________________________________________
dropout_2 (Dropout)          (None, 24, 24, 128)       0         
_________________________________________________________________
conv2d_3 (Conv2D)            (None, 24, 24, 256)       131328    
_________________________________________________________________
max_pooling2d_3 (MaxPooling2 (None, 12, 12, 256)       0         
_________________________________________________________________
dropout_3 (Dropout)          (None, 12, 12, 256)       0         
_________________________________________________________________
flatten_1 (Flatten)          (None, 36864)             0         
_________________________________________________________________
dense_1 (Dense)              (None, 256)               9437440   
_________________________________________________________________
dropout_4 (Dropout)          (None, 256)               0         
_________________________________________________________________
dense_2 (Dense)              (None, 150)               38550     
_________________________________________________________________
dropout_5 (Dropout)          (None, 150)               0         
_________________________________________________________________
dense_3 (Dense)              (None, 30)                4530      
=================================================================
Total params: 9,645,064
Trainable params: 9,645,064
Non-trainable params: 0
_________________________________________________________________

---

Step 6: Compile and Train the Model

After specifying your architecture, you'll need to compile and train the model to detect facial keypoints'

(IMPLEMENTATION) Compile and Train the Model

Use the compile method to configure the learning process. Experiment with your choice of optimizer; you may have some ideas about which will work best (SGD vs. RMSprop, etc), but take the time to empirically verify your theories.

Use the fit method to train the model. Break off a validation set by setting validation_split=0.2. Save the returned History object in the history variable.

Your model is required to attain a validation loss (measured as mean squared error) of at least XYZ. When you have finished training, save your model as an HDF5 file with file path my_model.h5.

from keras.optimizers import SGD, RMSprop, Adagrad, Adadelta, Adam, Adamax, Nadam
from keras.callbacks import ModelCheckpoint 

## TODO: Compile the model
hists = {}
epochs = 100
batch_size = 64

opts = ['SGD', 'RMSprop', 'Adagrad', 'Adadelta', 'Adam', 'Adamax', 'Nadam']

for opt in opts:
    print("Running Optimizer:", opt)
    model.compile(optimizer=opt, loss='mean_squared_error', metrics=['accuracy'])
    checkpointer = ModelCheckpoint(filepath=opt+'_weights_best.hdf5', verbose=1, save_best_only=True)

    ## TODO: Train the model
    hists[opt] = model.fit(X_train, y_train, validation_split=0.2, epochs=epochs, batch_size=batch_size,callbacks=[checkpointer], verbose=1, shuffle=True)

    ## TODO: Save the model as model.h5
    model.save(opt+'_model.h5')

Running Optimizer: SGD
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.1186 - acc: 0.1947Epoch 00000: val_loss improved from inf to 0.10797, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 7s - loss: 0.1181 - acc: 0.1986 - val_loss: 0.1080 - val_acc: 0.6963
Epoch 2/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0884 - acc: 0.2662Epoch 00001: val_loss improved from 0.10797 to 0.09326, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0883 - acc: 0.2669 - val_loss: 0.0933 - val_acc: 0.6963
Epoch 3/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0779 - acc: 0.2656Epoch 00002: val_loss improved from 0.09326 to 0.08488, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0777 - acc: 0.2664 - val_loss: 0.0849 - val_acc: 0.6963
Epoch 4/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0711 - acc: 0.2800Epoch 00003: val_loss improved from 0.08488 to 0.08005, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0709 - acc: 0.2792 - val_loss: 0.0801 - val_acc: 0.6963
Epoch 5/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0658 - acc: 0.3113Epoch 00004: val_loss improved from 0.08005 to 0.07572, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0656 - acc: 0.3102 - val_loss: 0.0757 - val_acc: 0.6963
Epoch 6/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0609 - acc: 0.2975Epoch 00005: val_loss improved from 0.07572 to 0.07248, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0609 - acc: 0.2979 - val_loss: 0.0725 - val_acc: 0.6963
Epoch 7/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0582 - acc: 0.3281Epoch 00006: val_loss improved from 0.07248 to 0.06949, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0580 - acc: 0.3254 - val_loss: 0.0695 - val_acc: 0.6963
Epoch 8/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0552 - acc: 0.3107Epoch 00007: val_loss improved from 0.06949 to 0.06675, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0552 - acc: 0.3078 - val_loss: 0.0668 - val_acc: 0.6963
Epoch 9/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0520 - acc: 0.3089Epoch 00008: val_loss improved from 0.06675 to 0.06478, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0521 - acc: 0.3102 - val_loss: 0.0648 - val_acc: 0.6963
Epoch 10/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0500 - acc: 0.3281Epoch 00009: val_loss improved from 0.06478 to 0.06237, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0500 - acc: 0.3312 - val_loss: 0.0624 - val_acc: 0.6963
Epoch 11/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0479 - acc: 0.3353Epoch 00010: val_loss improved from 0.06237 to 0.06042, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0478 - acc: 0.3411 - val_loss: 0.0604 - val_acc: 0.6963
Epoch 12/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0453 - acc: 0.3450Epoch 00011: val_loss improved from 0.06042 to 0.05839, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0453 - acc: 0.3440 - val_loss: 0.0584 - val_acc: 0.6963
Epoch 13/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0439 - acc: 0.3516Epoch 00012: val_loss improved from 0.05839 to 0.05785, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0439 - acc: 0.3522 - val_loss: 0.0578 - val_acc: 0.6963
Epoch 14/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0424 - acc: 0.3618Epoch 00013: val_loss improved from 0.05785 to 0.05614, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0424 - acc: 0.3639 - val_loss: 0.0561 - val_acc: 0.6963
Epoch 15/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0409 - acc: 0.3510Epoch 00014: val_loss improved from 0.05614 to 0.05533, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0409 - acc: 0.3516 - val_loss: 0.0553 - val_acc: 0.6963
Epoch 16/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0403 - acc: 0.3540Epoch 00015: val_loss improved from 0.05533 to 0.05456, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0402 - acc: 0.3581 - val_loss: 0.0546 - val_acc: 0.6963
Epoch 17/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0387 - acc: 0.3600Epoch 00016: val_loss improved from 0.05456 to 0.05305, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0386 - acc: 0.3592 - val_loss: 0.0530 - val_acc: 0.6963
Epoch 18/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0373 - acc: 0.3708Epoch 00017: val_loss improved from 0.05305 to 0.05180, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0374 - acc: 0.3732 - val_loss: 0.0518 - val_acc: 0.6963
Epoch 19/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0367 - acc: 0.3846Epoch 00018: val_loss improved from 0.05180 to 0.04964, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0367 - acc: 0.3814 - val_loss: 0.0496 - val_acc: 0.6963
Epoch 20/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0356 - acc: 0.3972Epoch 00019: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0355 - acc: 0.3972 - val_loss: 0.0500 - val_acc: 0.6963
Epoch 21/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0348 - acc: 0.4002Epoch 00020: val_loss improved from 0.04964 to 0.04889, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0347 - acc: 0.3984 - val_loss: 0.0489 - val_acc: 0.6963
Epoch 22/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0340 - acc: 0.3858Epoch 00021: val_loss improved from 0.04889 to 0.04733, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0340 - acc: 0.3838 - val_loss: 0.0473 - val_acc: 0.6963
Epoch 23/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0332 - acc: 0.4026Epoch 00022: val_loss improved from 0.04733 to 0.04680, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0333 - acc: 0.4042 - val_loss: 0.0468 - val_acc: 0.6963
Epoch 24/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0323 - acc: 0.3978Epoch 00023: val_loss improved from 0.04680 to 0.04584, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0323 - acc: 0.3966 - val_loss: 0.0458 - val_acc: 0.6963
Epoch 25/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0316 - acc: 0.4032Epoch 00024: val_loss improved from 0.04584 to 0.04545, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0315 - acc: 0.4030 - val_loss: 0.0455 - val_acc: 0.6963
Epoch 26/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0313 - acc: 0.4014Epoch 00025: val_loss improved from 0.04545 to 0.04462, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0313 - acc: 0.4042 - val_loss: 0.0446 - val_acc: 0.6963
Epoch 27/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0301 - acc: 0.4062Epoch 00026: val_loss improved from 0.04462 to 0.04377, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0302 - acc: 0.4048 - val_loss: 0.0438 - val_acc: 0.6963
Epoch 28/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0303 - acc: 0.4062Epoch 00027: val_loss improved from 0.04377 to 0.04233, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0304 - acc: 0.4054 - val_loss: 0.0423 - val_acc: 0.6963
Epoch 29/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0295 - acc: 0.4237Epoch 00028: val_loss improved from 0.04233 to 0.04221, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0295 - acc: 0.4206 - val_loss: 0.0422 - val_acc: 0.6963
Epoch 30/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0286 - acc: 0.4393Epoch 00029: val_loss improved from 0.04221 to 0.04146, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0285 - acc: 0.4357 - val_loss: 0.0415 - val_acc: 0.6963
Epoch 31/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0280 - acc: 0.4087Epoch 00030: val_loss improved from 0.04146 to 0.04056, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0280 - acc: 0.4106 - val_loss: 0.0406 - val_acc: 0.6963
Epoch 32/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0277 - acc: 0.4381Epoch 00031: val_loss improved from 0.04056 to 0.03986, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0277 - acc: 0.4416 - val_loss: 0.0399 - val_acc: 0.6963
Epoch 33/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0276 - acc: 0.3918Epoch 00032: val_loss improved from 0.03986 to 0.03929, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0275 - acc: 0.3879 - val_loss: 0.0393 - val_acc: 0.6963
Epoch 34/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0268 - acc: 0.4171Epoch 00033: val_loss improved from 0.03929 to 0.03914, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0268 - acc: 0.4141 - val_loss: 0.0391 - val_acc: 0.6963
Epoch 35/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0264 - acc: 0.4369Epoch 00034: val_loss improved from 0.03914 to 0.03860, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0264 - acc: 0.4387 - val_loss: 0.0386 - val_acc: 0.6963
Epoch 36/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0258 - acc: 0.4285Epoch 00035: val_loss improved from 0.03860 to 0.03770, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0258 - acc: 0.4270 - val_loss: 0.0377 - val_acc: 0.6963
Epoch 37/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0255 - acc: 0.4519Epoch 00036: val_loss improved from 0.03770 to 0.03703, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0255 - acc: 0.4498 - val_loss: 0.0370 - val_acc: 0.6963
Epoch 38/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0249 - acc: 0.4459Epoch 00037: val_loss improved from 0.03703 to 0.03646, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0249 - acc: 0.4422 - val_loss: 0.0365 - val_acc: 0.6963
Epoch 39/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0247 - acc: 0.4339Epoch 00038: val_loss improved from 0.03646 to 0.03594, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0247 - acc: 0.4346 - val_loss: 0.0359 - val_acc: 0.6963
Epoch 40/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0240 - acc: 0.4700Epoch 00039: val_loss improved from 0.03594 to 0.03553, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0241 - acc: 0.4702 - val_loss: 0.0355 - val_acc: 0.6963
Epoch 41/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0239 - acc: 0.4351Epoch 00040: val_loss improved from 0.03553 to 0.03470, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0240 - acc: 0.4340 - val_loss: 0.0347 - val_acc: 0.6963
Epoch 42/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0232 - acc: 0.4712Epoch 00041: val_loss improved from 0.03470 to 0.03397, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0232 - acc: 0.4743 - val_loss: 0.0340 - val_acc: 0.6963
Epoch 43/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0234 - acc: 0.4615Epoch 00042: val_loss improved from 0.03397 to 0.03395, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0234 - acc: 0.4614 - val_loss: 0.0340 - val_acc: 0.6963
Epoch 44/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0228 - acc: 0.4657Epoch 00043: val_loss improved from 0.03395 to 0.03344, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0229 - acc: 0.4667 - val_loss: 0.0334 - val_acc: 0.6963
Epoch 45/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0225 - acc: 0.4549Epoch 00044: val_loss improved from 0.03344 to 0.03285, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0225 - acc: 0.4556 - val_loss: 0.0328 - val_acc: 0.6963
Epoch 46/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0227 - acc: 0.4531Epoch 00045: val_loss improved from 0.03285 to 0.03218, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0227 - acc: 0.4521 - val_loss: 0.0322 - val_acc: 0.6963
Epoch 47/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0222 - acc: 0.4597Epoch 00046: val_loss improved from 0.03218 to 0.03195, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0222 - acc: 0.4574 - val_loss: 0.0320 - val_acc: 0.6963
Epoch 48/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0217 - acc: 0.4790Epoch 00047: val_loss improved from 0.03195 to 0.03164, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0217 - acc: 0.4766 - val_loss: 0.0316 - val_acc: 0.6963
Epoch 49/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0215 - acc: 0.4808Epoch 00048: val_loss improved from 0.03164 to 0.03114, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0215 - acc: 0.4801 - val_loss: 0.0311 - val_acc: 0.6963
Epoch 50/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0216 - acc: 0.4663Epoch 00049: val_loss improved from 0.03114 to 0.03038, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0216 - acc: 0.4655 - val_loss: 0.0304 - val_acc: 0.6963
Epoch 51/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0209 - acc: 0.5012Epoch 00050: val_loss improved from 0.03038 to 0.03005, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0209 - acc: 0.4977 - val_loss: 0.0300 - val_acc: 0.6963
Epoch 52/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0208 - acc: 0.4754Epoch 00051: val_loss improved from 0.03005 to 0.02964, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0208 - acc: 0.4725 - val_loss: 0.0296 - val_acc: 0.6963
Epoch 53/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0206 - acc: 0.4591Epoch 00052: val_loss improved from 0.02964 to 0.02925, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0206 - acc: 0.4579 - val_loss: 0.0293 - val_acc: 0.6963
Epoch 54/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0202 - acc: 0.4615Epoch 00053: val_loss improved from 0.02925 to 0.02900, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0202 - acc: 0.4550 - val_loss: 0.0290 - val_acc: 0.6963
Epoch 55/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0201 - acc: 0.4784Epoch 00054: val_loss improved from 0.02900 to 0.02831, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0201 - acc: 0.4766 - val_loss: 0.0283 - val_acc: 0.6963
Epoch 56/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0202 - acc: 0.4802Epoch 00055: val_loss improved from 0.02831 to 0.02811, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0202 - acc: 0.4813 - val_loss: 0.0281 - val_acc: 0.6963
Epoch 57/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0194 - acc: 0.4808Epoch 00056: val_loss improved from 0.02811 to 0.02746, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0194 - acc: 0.4813 - val_loss: 0.0275 - val_acc: 0.6963
Epoch 58/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0193 - acc: 0.5102Epoch 00057: val_loss improved from 0.02746 to 0.02714, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0193 - acc: 0.5093 - val_loss: 0.0271 - val_acc: 0.6963
Epoch 59/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0193 - acc: 0.5000Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0193 - acc: 0.4994 - val_loss: 0.0272 - val_acc: 0.6963
Epoch 60/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0192 - acc: 0.5030Epoch 00059: val_loss improved from 0.02714 to 0.02645, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0191 - acc: 0.5023 - val_loss: 0.0264 - val_acc: 0.6963
Epoch 61/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0189 - acc: 0.4820Epoch 00060: val_loss improved from 0.02645 to 0.02589, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0188 - acc: 0.4836 - val_loss: 0.0259 - val_acc: 0.6963
Epoch 62/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0183 - acc: 0.4808Epoch 00061: val_loss improved from 0.02589 to 0.02545, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0183 - acc: 0.4778 - val_loss: 0.0255 - val_acc: 0.6963
Epoch 63/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0185 - acc: 0.4838Epoch 00062: val_loss improved from 0.02545 to 0.02543, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0185 - acc: 0.4831 - val_loss: 0.0254 - val_acc: 0.6963
Epoch 64/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0183 - acc: 0.4946Epoch 00063: val_loss improved from 0.02543 to 0.02508, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0183 - acc: 0.4942 - val_loss: 0.0251 - val_acc: 0.6963
Epoch 65/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0179 - acc: 0.4796Epoch 00064: val_loss improved from 0.02508 to 0.02457, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0179 - acc: 0.4836 - val_loss: 0.0246 - val_acc: 0.6963
Epoch 66/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0179 - acc: 0.4946Epoch 00065: val_loss improved from 0.02457 to 0.02426, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0179 - acc: 0.4942 - val_loss: 0.0243 - val_acc: 0.6963
Epoch 67/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0175 - acc: 0.5096Epoch 00066: val_loss improved from 0.02426 to 0.02416, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0175 - acc: 0.5041 - val_loss: 0.0242 - val_acc: 0.6963
Epoch 68/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0177 - acc: 0.5054Epoch 00067: val_loss improved from 0.02416 to 0.02380, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0177 - acc: 0.5041 - val_loss: 0.0238 - val_acc: 0.6963
Epoch 69/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0175 - acc: 0.4922Epoch 00068: val_loss improved from 0.02380 to 0.02340, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0176 - acc: 0.4924 - val_loss: 0.0234 - val_acc: 0.6963
Epoch 70/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0174 - acc: 0.4778Epoch 00069: val_loss improved from 0.02340 to 0.02275, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0174 - acc: 0.4796 - val_loss: 0.0228 - val_acc: 0.6963
Epoch 71/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0173 - acc: 0.4850Epoch 00070: val_loss improved from 0.02275 to 0.02272, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0173 - acc: 0.4924 - val_loss: 0.0227 - val_acc: 0.6963
Epoch 72/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0170 - acc: 0.5090Epoch 00071: val_loss improved from 0.02272 to 0.02247, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0170 - acc: 0.5064 - val_loss: 0.0225 - val_acc: 0.6963
Epoch 73/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0165 - acc: 0.4964Epoch 00072: val_loss improved from 0.02247 to 0.02213, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0165 - acc: 0.5000 - val_loss: 0.0221 - val_acc: 0.6963
Epoch 74/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0165 - acc: 0.5240Epoch 00073: val_loss improved from 0.02213 to 0.02186, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0165 - acc: 0.5275 - val_loss: 0.0219 - val_acc: 0.6963
Epoch 75/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0166 - acc: 0.5090Epoch 00074: val_loss improved from 0.02186 to 0.02131, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0166 - acc: 0.5070 - val_loss: 0.0213 - val_acc: 0.6963
Epoch 76/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0165 - acc: 0.5144Epoch 00075: val_loss improved from 0.02131 to 0.02125, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0165 - acc: 0.5123 - val_loss: 0.0212 - val_acc: 0.6963
Epoch 77/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0163 - acc: 0.5186Epoch 00076: val_loss improved from 0.02125 to 0.02100, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0162 - acc: 0.5216 - val_loss: 0.0210 - val_acc: 0.6963
Epoch 78/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0158 - acc: 0.4964Epoch 00077: val_loss improved from 0.02100 to 0.02064, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0158 - acc: 0.4965 - val_loss: 0.0206 - val_acc: 0.6963
Epoch 79/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0157 - acc: 0.5114Epoch 00078: val_loss improved from 0.02064 to 0.02017, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0158 - acc: 0.5099 - val_loss: 0.0202 - val_acc: 0.6963
Epoch 80/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0159 - acc: 0.4964Epoch 00079: val_loss improved from 0.02017 to 0.01999, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0158 - acc: 0.4959 - val_loss: 0.0200 - val_acc: 0.6963
Epoch 81/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0156 - acc: 0.5024Epoch 00080: val_loss improved from 0.01999 to 0.01975, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0157 - acc: 0.5023 - val_loss: 0.0197 - val_acc: 0.6963
Epoch 82/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0156 - acc: 0.5162Epoch 00081: val_loss improved from 0.01975 to 0.01960, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0155 - acc: 0.5164 - val_loss: 0.0196 - val_acc: 0.6963
Epoch 83/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0155 - acc: 0.5126Epoch 00082: val_loss improved from 0.01960 to 0.01939, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0154 - acc: 0.5111 - val_loss: 0.0194 - val_acc: 0.6963
Epoch 84/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0152 - acc: 0.5282Epoch 00083: val_loss improved from 0.01939 to 0.01899, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0152 - acc: 0.5275 - val_loss: 0.0190 - val_acc: 0.6963
Epoch 85/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0154 - acc: 0.5054Epoch 00084: val_loss improved from 0.01899 to 0.01873, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0153 - acc: 0.5053 - val_loss: 0.0187 - val_acc: 0.6963
Epoch 86/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0153 - acc: 0.5252Epoch 00085: val_loss improved from 0.01873 to 0.01855, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0153 - acc: 0.5216 - val_loss: 0.0186 - val_acc: 0.6963
Epoch 87/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0149 - acc: 0.5343Epoch 00086: val_loss improved from 0.01855 to 0.01822, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0149 - acc: 0.5310 - val_loss: 0.0182 - val_acc: 0.6963
Epoch 88/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0151 - acc: 0.5084Epoch 00087: val_loss improved from 0.01822 to 0.01821, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0150 - acc: 0.5070 - val_loss: 0.0182 - val_acc: 0.6963
Epoch 89/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0148 - acc: 0.5162Epoch 00088: val_loss improved from 0.01821 to 0.01781, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0148 - acc: 0.5158 - val_loss: 0.0178 - val_acc: 0.6963
Epoch 90/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0147 - acc: 0.5156Epoch 00089: val_loss improved from 0.01781 to 0.01769, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0148 - acc: 0.5152 - val_loss: 0.0177 - val_acc: 0.6963
Epoch 91/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0145 - acc: 0.5126Epoch 00090: val_loss improved from 0.01769 to 0.01752, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0144 - acc: 0.5134 - val_loss: 0.0175 - val_acc: 0.6963
Epoch 92/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0146 - acc: 0.5463Epoch 00091: val_loss improved from 0.01752 to 0.01725, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0146 - acc: 0.5438 - val_loss: 0.0172 - val_acc: 0.6963
Epoch 93/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0146 - acc: 0.5024Epoch 00092: val_loss improved from 0.01725 to 0.01709, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0146 - acc: 0.5018 - val_loss: 0.0171 - val_acc: 0.6963
Epoch 94/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0143 - acc: 0.5421Epoch 00093: val_loss improved from 0.01709 to 0.01674, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0143 - acc: 0.5403 - val_loss: 0.0167 - val_acc: 0.6963
Epoch 95/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0143 - acc: 0.5294Epoch 00094: val_loss improved from 0.01674 to 0.01656, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0144 - acc: 0.5292 - val_loss: 0.0166 - val_acc: 0.6963
Epoch 96/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0140 - acc: 0.5541Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0140 - acc: 0.5537 - val_loss: 0.0166 - val_acc: 0.6963
Epoch 97/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0140 - acc: 0.5312Epoch 00096: val_loss improved from 0.01656 to 0.01622, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0140 - acc: 0.5304 - val_loss: 0.0162 - val_acc: 0.6963
Epoch 98/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0140 - acc: 0.5282Epoch 00097: val_loss improved from 0.01622 to 0.01603, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0140 - acc: 0.5292 - val_loss: 0.0160 - val_acc: 0.6963
Epoch 99/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0138 - acc: 0.5451Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0138 - acc: 0.5456 - val_loss: 0.0161 - val_acc: 0.6963
Epoch 100/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0136 - acc: 0.5240Epoch 00099: val_loss improved from 0.01603 to 0.01555, saving model to SGD_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0136 - acc: 0.5275 - val_loss: 0.0155 - val_acc: 0.6963
Running Optimizer: RMSprop
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0365 - acc: 0.5517Epoch 00000: val_loss improved from inf to 0.01197, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0358 - acc: 0.5508 - val_loss: 0.0120 - val_acc: 0.6963
Epoch 2/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0083 - acc: 0.6166Epoch 00001: val_loss improved from 0.01197 to 0.00471, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0084 - acc: 0.6180 - val_loss: 0.0047 - val_acc: 0.6963
Epoch 3/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0071 - acc: 0.6376Epoch 00002: val_loss improved from 0.00471 to 0.00442, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0070 - acc: 0.6384 - val_loss: 0.0044 - val_acc: 0.6963
Epoch 4/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0060 - acc: 0.6881Epoch 00003: val_loss improved from 0.00442 to 0.00431, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0060 - acc: 0.6863 - val_loss: 0.0043 - val_acc: 0.6963
Epoch 5/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0054 - acc: 0.6905Epoch 00004: val_loss improved from 0.00431 to 0.00424, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0054 - acc: 0.6893 - val_loss: 0.0042 - val_acc: 0.6963
Epoch 6/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0053 - acc: 0.7043Epoch 00005: val_loss improved from 0.00424 to 0.00420, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0052 - acc: 0.7062 - val_loss: 0.0042 - val_acc: 0.6963
Epoch 7/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0048 - acc: 0.6995Epoch 00006: val_loss improved from 0.00420 to 0.00383, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0048 - acc: 0.7027 - val_loss: 0.0038 - val_acc: 0.6963
Epoch 8/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.7097Epoch 00007: val_loss improved from 0.00383 to 0.00357, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0044 - acc: 0.7074 - val_loss: 0.0036 - val_acc: 0.6963
Epoch 9/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.7079Epoch 00008: val_loss improved from 0.00357 to 0.00314, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0040 - acc: 0.7074 - val_loss: 0.0031 - val_acc: 0.6963
Epoch 10/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0038 - acc: 0.7115Epoch 00009: val_loss improved from 0.00314 to 0.00258, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0038 - acc: 0.7068 - val_loss: 0.0026 - val_acc: 0.7009
Epoch 11/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0035 - acc: 0.7055Epoch 00010: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0035 - acc: 0.7074 - val_loss: 0.0029 - val_acc: 0.6963
Epoch 12/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0032 - acc: 0.7085Epoch 00011: val_loss improved from 0.00258 to 0.00232, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0032 - acc: 0.7097 - val_loss: 0.0023 - val_acc: 0.6986
Epoch 13/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0031 - acc: 0.7139Epoch 00012: val_loss improved from 0.00232 to 0.00221, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0031 - acc: 0.7155 - val_loss: 0.0022 - val_acc: 0.7150
Epoch 14/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0029 - acc: 0.7085Epoch 00013: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0029 - acc: 0.7079 - val_loss: 0.0024 - val_acc: 0.6963
Epoch 15/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0027 - acc: 0.7121Epoch 00014: val_loss improved from 0.00221 to 0.00193, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0027 - acc: 0.7114 - val_loss: 0.0019 - val_acc: 0.7033
Epoch 16/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0027 - acc: 0.7218Epoch 00015: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0027 - acc: 0.7214 - val_loss: 0.0019 - val_acc: 0.7056
Epoch 17/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0026 - acc: 0.7188Epoch 00016: val_loss improved from 0.00193 to 0.00191, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0026 - acc: 0.7185 - val_loss: 0.0019 - val_acc: 0.7103
Epoch 18/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0025 - acc: 0.7175Epoch 00017: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0025 - acc: 0.7179 - val_loss: 0.0019 - val_acc: 0.7126
Epoch 19/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0023 - acc: 0.7115Epoch 00018: val_loss improved from 0.00191 to 0.00174, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0023 - acc: 0.7132 - val_loss: 0.0017 - val_acc: 0.7243
Epoch 20/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0023 - acc: 0.7272Epoch 00019: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0023 - acc: 0.7296 - val_loss: 0.0022 - val_acc: 0.6986
Epoch 21/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0023 - acc: 0.7133Epoch 00020: val_loss improved from 0.00174 to 0.00170, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0023 - acc: 0.7132 - val_loss: 0.0017 - val_acc: 0.7196
Epoch 22/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0021 - acc: 0.7266Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0022 - acc: 0.7290 - val_loss: 0.0025 - val_acc: 0.7056
Epoch 23/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0021 - acc: 0.7380Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0021 - acc: 0.7377 - val_loss: 0.0017 - val_acc: 0.7079
Epoch 24/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0022 - acc: 0.7194Epoch 00023: val_loss improved from 0.00170 to 0.00156, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0021 - acc: 0.7220 - val_loss: 0.0016 - val_acc: 0.7173
Epoch 25/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0021 - acc: 0.7326Epoch 00024: val_loss improved from 0.00156 to 0.00154, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0021 - acc: 0.7319 - val_loss: 0.0015 - val_acc: 0.7243
Epoch 26/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0020 - acc: 0.7181Epoch 00025: val_loss improved from 0.00154 to 0.00149, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0020 - acc: 0.7190 - val_loss: 0.0015 - val_acc: 0.7220
Epoch 27/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0019 - acc: 0.7356Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0019 - acc: 0.7348 - val_loss: 0.0016 - val_acc: 0.7150
Epoch 28/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0019 - acc: 0.7224Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0019 - acc: 0.7237 - val_loss: 0.0019 - val_acc: 0.7407
Epoch 29/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0019 - acc: 0.7452Epoch 00028: val_loss improved from 0.00149 to 0.00141, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0019 - acc: 0.7471 - val_loss: 0.0014 - val_acc: 0.7266
Epoch 30/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0018 - acc: 0.7440Epoch 00029: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0018 - acc: 0.7418 - val_loss: 0.0015 - val_acc: 0.7290
Epoch 31/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0018 - acc: 0.7392Epoch 00030: val_loss improved from 0.00141 to 0.00139, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0018 - acc: 0.7389 - val_loss: 0.0014 - val_acc: 0.7313
Epoch 32/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0018 - acc: 0.7446Epoch 00031: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0018 - acc: 0.7412 - val_loss: 0.0017 - val_acc: 0.7243
Epoch 33/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0018 - acc: 0.7404Epoch 00032: val_loss improved from 0.00139 to 0.00136, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0018 - acc: 0.7412 - val_loss: 0.0014 - val_acc: 0.7453
Epoch 34/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0018 - acc: 0.7518Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0018 - acc: 0.7529 - val_loss: 0.0015 - val_acc: 0.7266
Epoch 35/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0017 - acc: 0.7506Epoch 00034: val_loss improved from 0.00136 to 0.00130, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0017 - acc: 0.7500 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 36/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0017 - acc: 0.7422Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0017 - acc: 0.7418 - val_loss: 0.0015 - val_acc: 0.7383
Epoch 37/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0017 - acc: 0.7494Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0017 - acc: 0.7500 - val_loss: 0.0014 - val_acc: 0.7453
Epoch 38/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0017 - acc: 0.7338Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0017 - acc: 0.7371 - val_loss: 0.0015 - val_acc: 0.7523
Epoch 39/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0016 - acc: 0.7428Epoch 00038: val_loss improved from 0.00130 to 0.00120, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0016 - acc: 0.7418 - val_loss: 0.0012 - val_acc: 0.7617
Epoch 40/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0016 - acc: 0.7494Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0016 - acc: 0.7488 - val_loss: 0.0013 - val_acc: 0.7430
Epoch 41/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0016 - acc: 0.7530Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0016 - acc: 0.7529 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 42/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0016 - acc: 0.7554Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0016 - acc: 0.7564 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 43/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7530Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0016 - acc: 0.7518 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 44/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7542Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0015 - acc: 0.7529 - val_loss: 0.0012 - val_acc: 0.7757
Epoch 45/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7572Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0015 - acc: 0.7558 - val_loss: 0.0012 - val_acc: 0.7570
Epoch 46/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7578Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0015 - acc: 0.7558 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 47/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7692Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0015 - acc: 0.7669 - val_loss: 0.0012 - val_acc: 0.7593
Epoch 48/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7458Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0015 - acc: 0.7471 - val_loss: 0.0012 - val_acc: 0.7710
Epoch 49/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7584Epoch 00048: val_loss improved from 0.00120 to 0.00120, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0015 - acc: 0.7593 - val_loss: 0.0012 - val_acc: 0.7687
Epoch 50/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7548Epoch 00049: val_loss improved from 0.00120 to 0.00119, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0014 - acc: 0.7564 - val_loss: 0.0012 - val_acc: 0.7687
Epoch 51/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7536Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0014 - acc: 0.7558 - val_loss: 0.0012 - val_acc: 0.7593
Epoch 52/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7620Epoch 00051: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0015 - acc: 0.7617 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 53/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7596Epoch 00052: val_loss improved from 0.00119 to 0.00116, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0014 - acc: 0.7605 - val_loss: 0.0012 - val_acc: 0.7383
Epoch 54/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7536Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0014 - acc: 0.7564 - val_loss: 0.0012 - val_acc: 0.7664
Epoch 55/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7656Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0014 - acc: 0.7640 - val_loss: 0.0012 - val_acc: 0.7734
Epoch 56/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7614Epoch 00055: val_loss improved from 0.00116 to 0.00113, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0014 - acc: 0.7617 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 57/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7548Epoch 00056: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0014 - acc: 0.7547 - val_loss: 0.0012 - val_acc: 0.7874
Epoch 58/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7644Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0014 - acc: 0.7652 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 59/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7656Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0014 - acc: 0.7652 - val_loss: 0.0012 - val_acc: 0.7687
Epoch 60/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7512Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7529 - val_loss: 0.0013 - val_acc: 0.7944
Epoch 61/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7674Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0014 - acc: 0.7669 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 62/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7728Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7704 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 63/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7620Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0014 - acc: 0.7634 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 64/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7668Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0013 - acc: 0.7629 - val_loss: 0.0012 - val_acc: 0.7617
Epoch 65/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7788Epoch 00064: val_loss improved from 0.00113 to 0.00112, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0014 - acc: 0.7786 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 66/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7566Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7593 - val_loss: 0.0012 - val_acc: 0.7617
Epoch 67/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7530Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7518 - val_loss: 0.0012 - val_acc: 0.7664
Epoch 68/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7788- ETA: 2s - loss: 0.0Epoch 00067: val_loss improved from 0.00112 to 0.00109, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7775 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 69/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7758Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7739 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 70/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7752Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7763 - val_loss: 0.0011 - val_acc: 0.7710
Epoch 71/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7782Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7775 - val_loss: 0.0012 - val_acc: 0.7874
Epoch 72/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7686Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0013 - acc: 0.7664 - val_loss: 0.0011 - val_acc: 0.7687
Epoch 73/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7885Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7868 - val_loss: 0.0012 - val_acc: 0.7780
Epoch 74/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7740Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7763 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 75/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7722Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0013 - acc: 0.7722 - val_loss: 0.0011 - val_acc: 0.7874
Epoch 76/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7806Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7821 - val_loss: 0.0012 - val_acc: 0.7734
Epoch 77/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7897Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0013 - acc: 0.7915 - val_loss: 0.0012 - val_acc: 0.7944
Epoch 78/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7710Epoch 00077: val_loss improved from 0.00109 to 0.00109, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7716 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 79/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7800Epoch 00078: val_loss improved from 0.00109 to 0.00109, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7804 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 80/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7861Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7839 - val_loss: 0.0012 - val_acc: 0.7780
Epoch 81/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7620Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0013 - acc: 0.7640 - val_loss: 0.0012 - val_acc: 0.7664
Epoch 82/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7819Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7833 - val_loss: 0.0012 - val_acc: 0.7780
Epoch 83/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7782Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0012 - acc: 0.7786 - val_loss: 0.0013 - val_acc: 0.7757
Epoch 84/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7770Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7751 - val_loss: 0.0012 - val_acc: 0.7757
Epoch 85/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7861Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0013 - acc: 0.7868 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 86/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7692Epoch 00085: val_loss improved from 0.00109 to 0.00108, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7704 - val_loss: 0.0011 - val_acc: 0.7827
Epoch 87/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7722Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0012 - acc: 0.7699 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 88/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7873Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7856 - val_loss: 0.0012 - val_acc: 0.7780
Epoch 89/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7867Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0012 - acc: 0.7874 - val_loss: 0.0012 - val_acc: 0.7827
Epoch 90/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7770Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7775 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 91/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7782Epoch 00090: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7786 - val_loss: 0.0015 - val_acc: 0.7734
Epoch 92/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7843Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0012 - acc: 0.7839 - val_loss: 0.0012 - val_acc: 0.7850
Epoch 93/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7740Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0012 - acc: 0.7763 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 94/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7963Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0012 - acc: 0.7973 - val_loss: 0.0012 - val_acc: 0.7804
Epoch 95/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7957Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7956 - val_loss: 0.0011 - val_acc: 0.7804
Epoch 96/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7903Epoch 00095: val_loss improved from 0.00108 to 0.00108, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7886 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 97/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7909Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7880 - val_loss: 0.0011 - val_acc: 0.7827
Epoch 98/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7855Epoch 00097: val_loss improved from 0.00108 to 0.00107, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7856 - val_loss: 0.0011 - val_acc: 0.7780
Epoch 99/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7819Epoch 00098: val_loss improved from 0.00107 to 0.00106, saving model to RMSprop_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7821 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 100/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7855Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7845 - val_loss: 0.0012 - val_acc: 0.7967
Running Optimizer: Adagrad
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.7145Epoch 00000: val_loss improved from inf to 0.00174, saving model to Adagrad_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0038 - acc: 0.7196 - val_loss: 0.0017 - val_acc: 0.7640
Epoch 2/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7518Epoch 00001: val_loss improved from 0.00174 to 0.00128, saving model to Adagrad_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0015 - acc: 0.7553 - val_loss: 0.0013 - val_acc: 0.7757
Epoch 3/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7776Epoch 00002: val_loss improved from 0.00128 to 0.00109, saving model to Adagrad_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7763 - val_loss: 0.0011 - val_acc: 0.7874
Epoch 4/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7698Epoch 00003: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0013 - acc: 0.7704 - val_loss: 0.0012 - val_acc: 0.7804
Epoch 5/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7764Epoch 00004: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7786 - val_loss: 0.0012 - val_acc: 0.7897
Epoch 6/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7831Epoch 00005: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0012 - acc: 0.7833 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 7/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7674Epoch 00006: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7699 - val_loss: 0.0012 - val_acc: 0.7804
Epoch 8/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7897Epoch 00007: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0012 - acc: 0.7897 - val_loss: 0.0012 - val_acc: 0.7850
Epoch 9/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7867Epoch 00008: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0012 - acc: 0.7874 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 10/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7812Epoch 00009: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7804 - val_loss: 0.0011 - val_acc: 0.7804
Epoch 11/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7752Epoch 00010: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7751 - val_loss: 0.0012 - val_acc: 0.7897
Epoch 12/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7794Epoch 00011: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7792 - val_loss: 0.0012 - val_acc: 0.7897
Epoch 13/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7698Epoch 00012: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7722 - val_loss: 0.0011 - val_acc: 0.7874
Epoch 14/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7969Epoch 00013: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7956 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 15/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7951Epoch 00014: val_loss improved from 0.00109 to 0.00107, saving model to Adagrad_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7909 - val_loss: 0.0011 - val_acc: 0.7827
Epoch 16/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7891Epoch 00015: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7909 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 17/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7794Epoch 00016: val_loss improved from 0.00107 to 0.00103, saving model to Adagrad_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7804 - val_loss: 0.0010 - val_acc: 0.7874
Epoch 18/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7891Epoch 00017: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7909 - val_loss: 0.0011 - val_acc: 0.7874
Epoch 19/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7909Epoch 00018: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7926 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 20/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8083Epoch 00019: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.8061 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 21/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7800Epoch 00020: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7804 - val_loss: 0.0011 - val_acc: 0.7874
Epoch 22/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7734Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7716 - val_loss: 0.0011 - val_acc: 0.7874
Epoch 23/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7849Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7850 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 24/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7867Epoch 00023: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7862 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 25/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7776Epoch 00024: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7786 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 26/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7794Epoch 00025: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7821 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 27/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7945Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7967 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 28/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7915Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7897 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 29/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7903Epoch 00028: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7891 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 30/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8017Epoch 00029: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8020 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 31/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7927Epoch 00030: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7926 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 32/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7885Epoch 00031: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7886 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 33/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7861Epoch 00032: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7880 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 34/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7885Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7897 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 35/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7939Epoch 00034: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7944 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 36/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7861Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7845 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 37/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7903Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7915 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 38/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7843Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7833 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 39/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7897Epoch 00038: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7921 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 40/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7849Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7850 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 41/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7939Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7944 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 42/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7831Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7850 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 43/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7975Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7956 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 44/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7951Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7967 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 45/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7915Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7897 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 46/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7981Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7961 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 47/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8005Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8032 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 48/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7915Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7880 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 49/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7831Epoch 00048: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7856 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 50/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7963Epoch 00049: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7967 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 51/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8035Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8043 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 52/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7855- ETA: 2s - loss: Epoch 00051: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7862 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 53/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7939Epoch 00052: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7944 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 54/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7915Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7903 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 55/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7909Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7932 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 56/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8029Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8055 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 57/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7800Epoch 00056: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7804 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 58/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7969Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7979 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 59/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7951Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7961 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 60/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7963Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7996 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 61/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8023Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8026 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 62/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8029Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8020 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 63/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7987Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7979 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 64/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7873Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7891 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 65/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7927Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7921 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 66/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7885Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7909 - val_loss: 0.0010 - val_acc: 0.7991
Epoch 67/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7921Epoch 00066: val_loss improved from 0.00103 to 0.00102, saving model to Adagrad_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7926 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 68/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7903Epoch 00067: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7909 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 69/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7969Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7938 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 70/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7969Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7996 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 71/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7957Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7973 - val_loss: 0.0010 - val_acc: 0.7967
Epoch 72/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8053Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8072 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 73/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7921Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7921 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 74/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8041Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8037 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 75/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7879Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7874 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 76/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7975Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7967 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 77/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8029Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8037 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 78/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7999Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7996 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 79/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7981Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8002 - val_loss: 0.0010 - val_acc: 0.7991
Epoch 80/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8059  Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8049 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 81/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8047Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8061 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 82/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9870e-04 - acc: 0.8047Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.9797e-04 - acc: 0.8032 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 83/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7915 Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7903 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 84/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8011Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.8014 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 85/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8017Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8043 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 86/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7927Epoch 00085: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7932 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 87/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7951Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7973 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 88/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8065Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8078 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 89/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8011Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8008 - val_loss: 0.0010 - val_acc: 0.7991
Epoch 90/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7945Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7944 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 91/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7951Epoch 00090: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7956 - val_loss: 0.0010 - val_acc: 0.7967
Epoch 92/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8035Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8037 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 93/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7837Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7850 - val_loss: 0.0010 - val_acc: 0.7991
Epoch 94/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7945Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7944 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 95/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7921Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7926 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 96/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9926e-04 - acc: 0.8149Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9760e-04 - acc: 0.8131 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 97/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7987Epoch 00096: val_loss improved from 0.00102 to 0.00101, saving model to Adagrad_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7967 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 98/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7993Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7985 - val_loss: 0.0010 - val_acc: 0.8131
Epoch 99/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7885  Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7891 - val_loss: 0.0011 - val_acc: 0.8154
Epoch 100/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7969Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7991 - val_loss: 0.0010 - val_acc: 0.8131
Running Optimizer: Adadelta
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8083Epoch 00000: val_loss improved from inf to 0.00106, saving model to Adadelta_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8067 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 2/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8125Epoch 00001: val_loss improved from 0.00106 to 0.00104, saving model to Adadelta_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8154 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 3/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.783 - ETA: 0s - loss: 0.0010 - acc: 0.7831Epoch 00002: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7839 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 4/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7981Epoch 00003: val_loss improved from 0.00104 to 0.00102, saving model to Adadelta_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7979 - val_loss: 0.0010 - val_acc: 0.8178
Epoch 5/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9784e-04 - acc: 0.7921Epoch 00004: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7921 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 6/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8068e-04 - acc: 0.8023Epoch 00005: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.8978e-04 - acc: 0.8049 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 7/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9522e-04 - acc: 0.7921Epoch 00006: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7909 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 8/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8029Epoch 00007: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8037 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 9/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9178e-04 - acc: 0.7969Epoch 00008: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7967 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 10/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8004e-04 - acc: 0.8071Epoch 00009: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.8204e-04 - acc: 0.8049 - val_loss: 0.0010 - val_acc: 0.8178
Epoch 11/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9951e-04 - acc: 0.8029Epoch 00010: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8020 - val_loss: 0.0010 - val_acc: 0.8131
Epoch 12/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7891   Epoch 00011: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7897 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 13/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7933Epoch 00012: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7938 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 14/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8107Epoch 00013: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8102 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 15/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8029    Epoch 00014: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8026 - val_loss: 0.0010 - val_acc: 0.8131
Epoch 16/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9364e-04 - acc: 0.7981Epoch 00015: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9497e-04 - acc: 0.7979 - val_loss: 0.0010 - val_acc: 0.8178
Epoch 17/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7873  Epoch 00016: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7868 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 18/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8149Epoch 00017: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8160 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 19/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9919e-04 - acc: 0.7873Epoch 00018: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9756e-04 - acc: 0.7874 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 20/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7981Epoch 00019: val_loss improved from 0.00102 to 0.00101, saving model to Adadelta_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8002 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 21/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8167Epoch 00020: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8148 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 22/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7969Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7973 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 23/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9150e-04 - acc: 0.7951Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9348e-04 - acc: 0.7944 - val_loss: 0.0011 - val_acc: 0.8178
Epoch 24/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7975Epoch 00023: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9948e-04 - acc: 0.7985 - val_loss: 0.0011 - val_acc: 0.8178
Epoch 25/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7993Epoch 00024: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7991 - val_loss: 0.0010 - val_acc: 0.8178
Epoch 26/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8065Epoch 00025: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8049 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 27/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.816 - ETA: 0s - loss: 0.0010 - acc: 0.8161Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9770e-04 - acc: 0.8178 - val_loss: 0.0010 - val_acc: 0.8131
Epoch 28/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7933Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7950 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 29/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7945Epoch 00028: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7932 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 30/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9868e-04 - acc: 0.8041Epoch 00029: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9737e-04 - acc: 0.8055 - val_loss: 0.0010 - val_acc: 0.8131
Epoch 31/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7993Epoch 00030: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8008 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 32/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7957Epoch 00031: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7991 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 33/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8077Epoch 00032: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8078 - val_loss: 0.0010 - val_acc: 0.8178
Epoch 34/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7957Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7956 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 35/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8131Epoch 00034: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8137 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 36/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9859e-04 - acc: 0.8035Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9851e-04 - acc: 0.8061 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 37/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7903Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7915 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 38/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9401e-04 - acc: 0.7969Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9193e-04 - acc: 0.7973 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 39/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7957Epoch 00038: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7956 - val_loss: 0.0010 - val_acc: 0.8131
Epoch 40/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7927Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7944 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 41/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8155Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8119 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 42/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7963Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7950 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 43/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9506e-04 - acc: 0.8017Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9346e-04 - acc: 0.8014 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 44/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8209  Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9931e-04 - acc: 0.8178 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 45/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7981Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7979 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 46/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8029Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8043 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 47/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7969    Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7967 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 48/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8929e-04 - acc: 0.8083Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.8850e-04 - acc: 0.8078 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 49/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8131Epoch 00048: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8143 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 50/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9653e-04 - acc: 0.7939Epoch 00049: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.9754e-04 - acc: 0.7956 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 51/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8167Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8201 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 52/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8107Epoch 00051: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8096 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 53/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8065    Epoch 00052: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8084 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 54/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8119  Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8119 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 55/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7993Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7996 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 56/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7951 Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7938 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 57/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9734e-04 - acc: 0.7939Epoch 00056: val_loss improved from 0.00101 to 0.00101, saving model to Adadelta_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7915 - val_loss: 0.0010 - val_acc: 0.8131
Epoch 58/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9998e-04 - acc: 0.8041Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9581e-04 - acc: 0.8061 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 59/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7957Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7979 - val_loss: 0.0010 - val_acc: 0.8154
Epoch 60/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7885Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7897 - val_loss: 0.0010 - val_acc: 0.8154
Epoch 61/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7933Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7950 - val_loss: 0.0010 - val_acc: 0.8201
Epoch 62/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7873Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7868 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 63/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7963Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7950 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 64/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7975Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7961 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 65/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8011Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8026 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 66/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8769e-04 - acc: 0.8173Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.8588e-04 - acc: 0.8178 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 67/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7909 Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7921 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 68/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.7832e-04 - acc: 0.7957Epoch 00067: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.7655e-04 - acc: 0.7961 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 69/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8089Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8096 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 70/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8005Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8002 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 71/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7867Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7862 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 72/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8958e-04 - acc: 0.8011Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9598e-04 - acc: 0.8032 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 73/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9539e-04 - acc: 0.7987Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9428e-04 - acc: 0.7985 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 74/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7891Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7874 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 75/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7915  Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7897 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 76/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8095Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8090 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 77/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8023Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8037 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 78/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7867Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7868 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 79/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7969Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7991 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 80/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7921 Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7932 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 81/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8005Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8008 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 82/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8035   Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8043 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 83/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8047Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8049 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 84/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7825Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7821 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 85/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8113Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8107 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 86/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7794Epoch 00085: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7815 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 87/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8155Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8148 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 88/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7993Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8008 - val_loss: 0.0010 - val_acc: 0.7991
Epoch 89/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8053Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8061 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 90/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8017Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8002 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 91/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9396e-04 - acc: 0.8053Epoch 00090: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9813e-04 - acc: 0.8067 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 92/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8531e-04 - acc: 0.8143Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.8886e-04 - acc: 0.8148 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 93/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9054e-04 - acc: 0.8065Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.9599e-04 - acc: 0.8078 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 94/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7975    Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7985 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 95/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8011Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8014 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 96/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8011Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8014 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 97/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8017    Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8026 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 98/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8036e-04 - acc: 0.8125Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.8034e-04 - acc: 0.8125 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 99/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8357e-04 - acc: 0.8113Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.8910e-04 - acc: 0.8125 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 100/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7909- ETA: 3s - lossEpoch 00099: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7926 - val_loss: 0.0010 - val_acc: 0.8084
Running Optimizer: Adam
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7794Epoch 00000: val_loss improved from inf to 0.00116, saving model to Adam_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7804 - val_loss: 0.0012 - val_acc: 0.7991
Epoch 2/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7993Epoch 00001: val_loss improved from 0.00116 to 0.00109, saving model to Adam_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7967 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 3/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7861Epoch 00002: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7862 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 4/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7993Epoch 00003: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7979 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 5/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7891Epoch 00004: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7868 - val_loss: 0.0012 - val_acc: 0.8061
Epoch 6/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7849Epoch 00005: val_loss improved from 0.00109 to 0.00106, saving model to Adam_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7886 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 7/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7843Epoch 00006: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7821 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 8/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8089Epoch 00007: val_loss improved from 0.00106 to 0.00103, saving model to Adam_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8067 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 9/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7975Epoch 00008: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7961 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 10/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7897Epoch 00009: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7915 - val_loss: 0.0010 - val_acc: 0.8154
Epoch 11/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7987Epoch 00010: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7991 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 12/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7927Epoch 00011: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7932 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 13/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7897Epoch 00012: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7915 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 14/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7999Epoch 00013: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8008 - val_loss: 0.0011 - val_acc: 0.7874
Epoch 15/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7933Epoch 00014: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7944 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 16/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7903Epoch 00015: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7897 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 17/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7951Epoch 00016: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7950 - val_loss: 0.0011 - val_acc: 0.8154
Epoch 18/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8113Epoch 00017: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.8107 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 19/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7903Epoch 00018: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7915 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 20/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7915Epoch 00019: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7926 - val_loss: 0.0012 - val_acc: 0.7991
Epoch 21/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7933Epoch 00020: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7921 - val_loss: 0.0011 - val_acc: 0.8154
Epoch 22/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7855Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7874 - val_loss: 0.0011 - val_acc: 0.7827
Epoch 23/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7945Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7938 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 24/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8089Epoch 00023: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8055 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 25/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7927Epoch 00024: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7938 - val_loss: 0.0011 - val_acc: 0.8178
Epoch 26/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8089Epoch 00025: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.8084 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 27/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8035Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.8067 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 28/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7909Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7921 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 29/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7951Epoch 00028: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7956 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 30/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7915Epoch 00029: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7932 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 31/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8113Epoch 00030: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8107 - val_loss: 0.0012 - val_acc: 0.7921
Epoch 32/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8131Epoch 00031: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8102 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 33/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7951Epoch 00032: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7973 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 34/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7831Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7827 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 35/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8041Epoch 00034: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8055 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 36/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8023Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8037 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 37/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8113Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8072 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 38/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7957Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7967 - val_loss: 0.0012 - val_acc: 0.7967
Epoch 39/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8023Epoch 00038: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8026 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 40/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8029Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 41/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7999Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7996 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 42/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8083Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8090 - val_loss: 0.0011 - val_acc: 0.8178
Epoch 43/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8011Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 44/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8023Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8043 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 45/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8053Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8043 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 46/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7999Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7991 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 47/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7963Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7956 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 48/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8017Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7996 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 49/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8131Epoch 00048: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8154 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 50/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7993Epoch 00049: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8020 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 51/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8065Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8072 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 52/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7999Epoch 00051: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7991 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 53/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7981Epoch 00052: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7991 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 54/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8131Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8125 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 55/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8083Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8037 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 56/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8089Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8078 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 57/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8017Epoch 00056: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 58/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8083Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.8072 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 59/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8065Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8026 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 60/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8047Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8026 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 61/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8149Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8148 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 62/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8101Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8107 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 63/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8053Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8055 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 64/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8137Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8131 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 65/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7993Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7973 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 66/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7957Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.7979 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 67/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8035Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8055 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 68/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8017Epoch 00067: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 69/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8005Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 70/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8089Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8078 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 71/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7849Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7868 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 72/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9748e-04 - acc: 0.8059Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.9480e-04 - acc: 0.8049 - val_loss: 0.0010 - val_acc: 0.7991
Epoch 73/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8083Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8055 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 74/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8035Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8043 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 75/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8107Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8125 - val_loss: 0.0011 - val_acc: 0.8224
Epoch 76/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8101Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8102 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 77/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7987Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7956 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 78/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8185Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8183 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 79/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8137Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8125 - val_loss: 0.0011 - val_acc: 0.8154
Epoch 80/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8077Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8072 - val_loss: 0.0011 - val_acc: 0.8248
Epoch 81/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8113Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8119 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 82/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8023Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 83/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7933Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7886 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 84/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8620e-04 - acc: 0.8131Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.8445e-04 - acc: 0.8143 - val_loss: 0.0011 - val_acc: 0.8154
Epoch 85/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7933Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7944 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 86/100
1664/1712 [============================>.] - ETA: 0s - loss: 1.0000e-03 - acc: 0.7999Epoch 00085: val_loss improved from 0.00103 to 0.00102, saving model to Adam_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 9.9945e-04 - acc: 0.7991 - val_loss: 0.0010 - val_acc: 0.7921
Epoch 87/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8107Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0010 - acc: 0.8084 - val_loss: 0.0011 - val_acc: 0.7827
Epoch 88/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8191Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8143 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 89/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8113Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8131 - val_loss: 0.0011 - val_acc: 0.7757
Epoch 90/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8095  Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8125 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 91/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7981Epoch 00090: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8008 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 92/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8972e-04 - acc: 0.8029Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.8952e-04 - acc: 0.8008 - val_loss: 0.0010 - val_acc: 0.7897
Epoch 93/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8083    Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8090 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 94/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9599e-04 - acc: 0.8047Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.9391e-04 - acc: 0.8037 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 95/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9427e-04 - acc: 0.8113Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9378e-04 - acc: 0.8137 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 96/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8005Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.9871e-04 - acc: 0.8032 - val_loss: 0.0012 - val_acc: 0.7850
Epoch 97/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9780e-04 - acc: 0.8101Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9814e-04 - acc: 0.8131 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 98/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.9283e-04 - acc: 0.8185Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.9381e-04 - acc: 0.8148 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 99/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8065Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8067 - val_loss: 0.0011 - val_acc: 0.7734
Epoch 100/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8047Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8049 - val_loss: 0.0011 - val_acc: 0.7827
Running Optimizer: Adamax
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8119Epoch 00000: val_loss improved from inf to 0.00114, saving model to Adamax_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8137 - val_loss: 0.0011 - val_acc: 0.7827
Epoch 2/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8161Epoch 00001: val_loss improved from 0.00114 to 0.00108, saving model to Adamax_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8154 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 3/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8407e-04 - acc: 0.8179Epoch 00002: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.8673e-04 - acc: 0.8178 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 4/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.7852e-04 - acc: 0.8011Epoch 00003: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.7872e-04 - acc: 0.8008 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 5/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.6293e-04 - acc: 0.8107Epoch 00004: val_loss improved from 0.00108 to 0.00103, saving model to Adamax_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 9.6590e-04 - acc: 0.8102 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 6/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.7216e-04 - acc: 0.8101Epoch 00005: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.7206e-04 - acc: 0.8084 - val_loss: 0.0010 - val_acc: 0.7991
Epoch 7/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.6134e-04 - acc: 0.8023Epoch 00006: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.5919e-04 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.8201
Epoch 8/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.6044e-04 - acc: 0.8071Epoch 00007: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.5925e-04 - acc: 0.8067 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 9/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.7181e-04 - acc: 0.8107Epoch 00008: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.7059e-04 - acc: 0.8102 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 10/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4585e-04 - acc: 0.8197Epoch 00009: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.4692e-04 - acc: 0.8201 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 11/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5797e-04 - acc: 0.8125Epoch 00010: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.5628e-04 - acc: 0.8137 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 12/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.6651e-04 - acc: 0.8023Epoch 00011: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.6548e-04 - acc: 0.8026 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 13/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5926e-04 - acc: 0.8155Epoch 00012: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.5636e-04 - acc: 0.8148 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 14/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.6750e-04 - acc: 0.8125Epoch 00013: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.7370e-04 - acc: 0.8107 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 15/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3917e-04 - acc: 0.8197Epoch 00014: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.3889e-04 - acc: 0.8218 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 16/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.7066e-04 - acc: 0.8089Epoch 00015: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.6627e-04 - acc: 0.8102 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 17/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5764e-04 - acc: 0.8035Epoch 00016: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.5679e-04 - acc: 0.8043 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 18/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3546e-04 - acc: 0.8191Epoch 00017: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.3379e-04 - acc: 0.8189 - val_loss: 0.0010 - val_acc: 0.7967
Epoch 19/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4890e-04 - acc: 0.8131Epoch 00018: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.5020e-04 - acc: 0.8131 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 20/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4482e-04 - acc: 0.8101Epoch 00019: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.4704e-04 - acc: 0.8107 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 21/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.6405e-04 - acc: 0.8023Epoch 00020: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.6598e-04 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 22/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5968e-04 - acc: 0.7987Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.5962e-04 - acc: 0.7991 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 23/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4755e-04 - acc: 0.8119Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.4665e-04 - acc: 0.8113 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 24/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4264e-04 - acc: 0.8155Epoch 00023: val_loss improved from 0.00103 to 0.00102, saving model to Adamax_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 9.4585e-04 - acc: 0.8160 - val_loss: 0.0010 - val_acc: 0.8178
Epoch 25/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5960e-04 - acc: 0.8107Epoch 00024: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.5838e-04 - acc: 0.8107 - val_loss: 0.0010 - val_acc: 0.8061
Epoch 26/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4978e-04 - acc: 0.8251Epoch 00025: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.5018e-04 - acc: 0.8248 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 27/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.6400e-04 - acc: 0.8191Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.6584e-04 - acc: 0.8189 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 28/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.6990e-04 - acc: 0.8047Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.6765e-04 - acc: 0.8049 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 29/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5087e-04 - acc: 0.8185Epoch 00028: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.5036e-04 - acc: 0.8178 - val_loss: 0.0010 - val_acc: 0.7991
Epoch 30/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5968e-04 - acc: 0.8083Epoch 00029: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.5173e-04 - acc: 0.8096 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 31/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3800e-04 - acc: 0.8209Epoch 00030: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.3328e-04 - acc: 0.8242 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 32/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4446e-04 - acc: 0.8029Epoch 00031: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.4813e-04 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 33/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.8384e-04 - acc: 0.8125Epoch 00032: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.8419e-04 - acc: 0.8090 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 34/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.6992e-04 - acc: 0.8221Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.6489e-04 - acc: 0.8242 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 35/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2968e-04 - acc: 0.8143Epoch 00034: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3110e-04 - acc: 0.8148 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 36/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.6360e-04 - acc: 0.8029Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.6034e-04 - acc: 0.8037 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 37/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4673e-04 - acc: 0.8293Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.4281e-04 - acc: 0.8300 - val_loss: 0.0010 - val_acc: 0.7944
Epoch 38/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2875e-04 - acc: 0.8221Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3820e-04 - acc: 0.8213 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 39/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4522e-04 - acc: 0.8101Epoch 00038: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.4064e-04 - acc: 0.8090 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 40/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4869e-04 - acc: 0.8149Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.4981e-04 - acc: 0.8148 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 41/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3616e-04 - acc: 0.8137Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.4184e-04 - acc: 0.8137 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 42/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2618e-04 - acc: 0.8137Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2877e-04 - acc: 0.8137 - val_loss: 0.0010 - val_acc: 0.7991
Epoch 43/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4260e-04 - acc: 0.8185Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.4129e-04 - acc: 0.8183 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 44/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2998e-04 - acc: 0.8095Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2947e-04 - acc: 0.8090 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 45/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5724e-04 - acc: 0.8131Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.5642e-04 - acc: 0.8125 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 46/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3368e-04 - acc: 0.8227Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.3618e-04 - acc: 0.8218 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 47/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3830e-04 - acc: 0.8131Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.3568e-04 - acc: 0.8143 - val_loss: 0.0011 - val_acc: 0.7874
Epoch 48/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3784e-04 - acc: 0.8239Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3612e-04 - acc: 0.8236 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 49/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3118e-04 - acc: 0.8161Epoch 00048: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2562e-04 - acc: 0.8189 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 50/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3497e-04 - acc: 0.8233Epoch 00049: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3829e-04 - acc: 0.8230 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 51/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5112e-04 - acc: 0.8065Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.5434e-04 - acc: 0.8090 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 52/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3530e-04 - acc: 0.8137Epoch 00051: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.3920e-04 - acc: 0.8107 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 53/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4495e-04 - acc: 0.8065Epoch 00052: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.4564e-04 - acc: 0.8061 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 54/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5540e-04 - acc: 0.8161Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.5253e-04 - acc: 0.8183 - val_loss: 0.0010 - val_acc: 0.8107
Epoch 55/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5618e-04 - acc: 0.8023Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.5750e-04 - acc: 0.8002 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 56/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4601e-04 - acc: 0.8095Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.4137e-04 - acc: 0.8102 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 57/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3347e-04 - acc: 0.8065Epoch 00056: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3939e-04 - acc: 0.8061 - val_loss: 0.0011 - val_acc: 0.8154
Epoch 58/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.1748e-04 - acc: 0.8209Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2110e-04 - acc: 0.8189 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 59/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2360e-04 - acc: 0.8185Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2730e-04 - acc: 0.8189 - val_loss: 0.0011 - val_acc: 0.8154
Epoch 60/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3834e-04 - acc: 0.8305Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.4200e-04 - acc: 0.8318 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 61/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4293e-04 - acc: 0.8131Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3923e-04 - acc: 0.8107 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 62/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5522e-04 - acc: 0.7999Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.5923e-04 - acc: 0.8008 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 63/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3897e-04 - acc: 0.8047Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3424e-04 - acc: 0.8078 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 64/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4778e-04 - acc: 0.8245Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.4805e-04 - acc: 0.8224 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 65/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4018e-04 - acc: 0.8239Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.4519e-04 - acc: 0.8207 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 66/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2805e-04 - acc: 0.8209Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.2956e-04 - acc: 0.8183 - val_loss: 0.0010 - val_acc: 0.8084
Epoch 67/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3113e-04 - acc: 0.8257Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.2698e-04 - acc: 0.8259 - val_loss: 0.0010 - val_acc: 0.7897
Epoch 68/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3222e-04 - acc: 0.8137Epoch 00067: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3318e-04 - acc: 0.8143 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 69/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3526e-04 - acc: 0.8071Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3298e-04 - acc: 0.8102 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 70/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.1916e-04 - acc: 0.8143Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.2155e-04 - acc: 0.8143 - val_loss: 0.0010 - val_acc: 0.8014
Epoch 71/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2253e-04 - acc: 0.8059Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2653e-04 - acc: 0.8078 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 72/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3057e-04 - acc: 0.8119Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3326e-04 - acc: 0.8125 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 73/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.5040e-04 - acc: 0.8251Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.4901e-04 - acc: 0.8259 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 74/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4175e-04 - acc: 0.8161Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.4096e-04 - acc: 0.8143 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 75/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3608e-04 - acc: 0.8071Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3936e-04 - acc: 0.8090 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 76/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.1272e-04 - acc: 0.8119Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.1384e-04 - acc: 0.8102 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 77/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3731e-04 - acc: 0.8233Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.3771e-04 - acc: 0.8224 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 78/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.1669e-04 - acc: 0.8101Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2132e-04 - acc: 0.8119 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 79/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3549e-04 - acc: 0.8101Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.3632e-04 - acc: 0.8084 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 80/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3220e-04 - acc: 0.8275Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3964e-04 - acc: 0.8218 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 81/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2187e-04 - acc: 0.8125Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2599e-04 - acc: 0.8119 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 82/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4953e-04 - acc: 0.8155Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.5370e-04 - acc: 0.8166 - val_loss: 0.0010 - val_acc: 0.8131
Epoch 83/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.1899e-04 - acc: 0.8287Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.2296e-04 - acc: 0.8277 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 84/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.1647e-04 - acc: 0.8071Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.1690e-04 - acc: 0.8061 - val_loss: 0.0011 - val_acc: 0.8154
Epoch 85/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2583e-04 - acc: 0.8287Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3025e-04 - acc: 0.8283 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 86/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2725e-04 - acc: 0.8215Epoch 00085: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.2783e-04 - acc: 0.8213 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 87/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2294e-04 - acc: 0.8035Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.2204e-04 - acc: 0.8055 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 88/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.1393e-04 - acc: 0.8203Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.1472e-04 - acc: 0.8201 - val_loss: 0.0011 - val_acc: 0.8061
Epoch 89/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2391e-04 - acc: 0.7951Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.3051e-04 - acc: 0.7967 - val_loss: 0.0010 - val_acc: 0.8201
Epoch 90/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2018e-04 - acc: 0.8245Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2308e-04 - acc: 0.8248 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 91/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2772e-04 - acc: 0.8209Epoch 00090: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3036e-04 - acc: 0.8201 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 92/100
1664/1712 [============================>.] - ETA: 0s - loss: 8.9026e-04 - acc: 0.8179Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 8.9821e-04 - acc: 0.8154 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 93/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2362e-04 - acc: 0.8281Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2992e-04 - acc: 0.8242 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 94/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3184e-04 - acc: 0.8287Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.3169e-04 - acc: 0.8265 - val_loss: 0.0010 - val_acc: 0.8037
Epoch 95/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2566e-04 - acc: 0.8065Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2776e-04 - acc: 0.8072 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 96/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.0629e-04 - acc: 0.8113Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.0990e-04 - acc: 0.8072 - val_loss: 0.0011 - val_acc: 0.8107
Epoch 97/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2113e-04 - acc: 0.8113Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.2151e-04 - acc: 0.8096 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 98/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.3027e-04 - acc: 0.8143Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 9.2778e-04 - acc: 0.8160 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 99/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.4354e-04 - acc: 0.8161Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.4330e-04 - acc: 0.8154 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 100/100
1664/1712 [============================>.] - ETA: 0s - loss: 9.2893e-04 - acc: 0.8245Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 9.2738e-04 - acc: 0.8242 - val_loss: 0.0011 - val_acc: 0.8037
Running Optimizer: Nadam
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8125Epoch 00000: val_loss improved from inf to 0.00116, saving model to Nadam_weights_best.hdf5
1712/1712 [==============================] - 5s - loss: 0.0010 - acc: 0.8113 - val_loss: 0.0012 - val_acc: 0.7967
Epoch 2/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8065- ETA: 0s - loss: 0.0011 - acc: 0.80Epoch 00001: val_loss improved from 0.00116 to 0.00112, saving model to Nadam_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8084 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 3/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8023Epoch 00002: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8032 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 4/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7993Epoch 00003: val_loss improved from 0.00112 to 0.00111, saving model to Nadam_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7973 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 5/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7999Epoch 00004: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 6/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8119Epoch 00005: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8119 - val_loss: 0.0012 - val_acc: 0.7991
Epoch 7/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8065Epoch 00006: val_loss improved from 0.00111 to 0.00106, saving model to Nadam_weights_best.hdf5
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8055 - val_loss: 0.0011 - val_acc: 0.8201
Epoch 8/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7861Epoch 00007: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0013 - acc: 0.7839 - val_loss: 0.0026 - val_acc: 0.7266
Epoch 9/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7746Epoch 00008: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0014 - acc: 0.7722 - val_loss: 0.0012 - val_acc: 0.8084
Epoch 10/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7951Epoch 00009: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.7938 - val_loss: 0.0011 - val_acc: 0.8084
Epoch 11/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7915Epoch 00010: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0013 - acc: 0.7915 - val_loss: 0.0013 - val_acc: 0.7897
Epoch 12/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7975Epoch 00011: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7979 - val_loss: 0.0012 - val_acc: 0.7897
Epoch 13/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7933Epoch 00012: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7950 - val_loss: 0.0012 - val_acc: 0.7897
Epoch 14/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7963Epoch 00013: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7961 - val_loss: 0.0012 - val_acc: 0.7944
Epoch 15/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8185Epoch 00014: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8183 - val_loss: 0.0012 - val_acc: 0.8037
Epoch 16/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8029Epoch 00015: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8014 - val_loss: 0.0012 - val_acc: 0.8061
Epoch 17/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8011Epoch 00016: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8008 - val_loss: 0.0013 - val_acc: 0.7991
Epoch 18/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8023Epoch 00017: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8026 - val_loss: 0.0012 - val_acc: 0.7944
Epoch 19/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8107Epoch 00018: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8131 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 20/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8005Epoch 00019: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8002 - val_loss: 0.0012 - val_acc: 0.7850
Epoch 21/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7921Epoch 00020: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7915 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 22/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8137Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8166 - val_loss: 0.0012 - val_acc: 0.7827
Epoch 23/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8011Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8002 - val_loss: 0.0012 - val_acc: 0.7827
Epoch 24/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7951Epoch 00023: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7944 - val_loss: 0.0012 - val_acc: 0.7991
Epoch 25/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8047Epoch 00024: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8061 - val_loss: 0.0012 - val_acc: 0.7944
Epoch 26/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8095Epoch 00025: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8107 - val_loss: 0.0014 - val_acc: 0.7500
Epoch 27/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7945Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7961 - val_loss: 0.0011 - val_acc: 0.7944
Epoch 28/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8059Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8055 - val_loss: 0.0012 - val_acc: 0.8014
Epoch 29/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8083- ETA: 0s - loss: 0.0010 - acc: Epoch 00028: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8055 - val_loss: 0.0011 - val_acc: 0.7780
Epoch 30/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8041Epoch 00029: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8055 - val_loss: 0.0012 - val_acc: 0.7850
Epoch 31/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8005Epoch 00030: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7996 - val_loss: 0.0011 - val_acc: 0.8037
Epoch 32/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7963Epoch 00031: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7979 - val_loss: 0.0011 - val_acc: 0.8014
Epoch 33/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7969Epoch 00032: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7967 - val_loss: 0.0015 - val_acc: 0.7453
Epoch 34/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8059Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8055 - val_loss: 0.0012 - val_acc: 0.7780
Epoch 35/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7987Epoch 00034: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7979 - val_loss: 0.0012 - val_acc: 0.7967
Epoch 36/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8125Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8125 - val_loss: 0.0012 - val_acc: 0.7897
Epoch 37/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7957Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7979 - val_loss: 0.0012 - val_acc: 0.7827
Epoch 38/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7927Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7932 - val_loss: 0.0011 - val_acc: 0.7687
Epoch 39/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7951Epoch 00038: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7938 - val_loss: 0.0012 - val_acc: 0.8084
Epoch 40/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8119Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8107 - val_loss: 0.0012 - val_acc: 0.8014
Epoch 41/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8107Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8096 - val_loss: 0.0012 - val_acc: 0.7967
Epoch 42/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7999Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8008 - val_loss: 0.0012 - val_acc: 0.8107
Epoch 43/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7993Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7996 - val_loss: 0.0012 - val_acc: 0.7944
Epoch 44/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8113Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8096 - val_loss: 0.0011 - val_acc: 0.7921
Epoch 45/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0016 - acc: 0.7740Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0016 - acc: 0.7734 - val_loss: 0.0012 - val_acc: 0.8037
Epoch 46/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.8035Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0012 - acc: 0.8008 - val_loss: 0.0013 - val_acc: 0.7897
Epoch 47/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7975Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7985 - val_loss: 0.0012 - val_acc: 0.7850
Epoch 48/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7963Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7979 - val_loss: 0.0012 - val_acc: 0.7897
Epoch 49/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7933Epoch 00048: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7938 - val_loss: 0.0011 - val_acc: 0.8131
Epoch 50/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8041Epoch 00049: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8049 - val_loss: 0.0012 - val_acc: 0.7944
Epoch 51/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7867Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7903 - val_loss: 0.0013 - val_acc: 0.7944
Epoch 52/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8005Epoch 00051: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8037 - val_loss: 0.0012 - val_acc: 0.8037
Epoch 53/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8053Epoch 00052: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.8043 - val_loss: 0.0012 - val_acc: 0.7921
Epoch 54/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7981Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.7973 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 55/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8197Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8189 - val_loss: 0.0012 - val_acc: 0.7874
Epoch 56/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8227Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8207 - val_loss: 0.0012 - val_acc: 0.7991
Epoch 57/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8023Epoch 00056: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8020 - val_loss: 0.0012 - val_acc: 0.7874
Epoch 58/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7987Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7985 - val_loss: 0.0014 - val_acc: 0.7477
Epoch 59/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7915Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7921 - val_loss: 0.0012 - val_acc: 0.7874
Epoch 60/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7915Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7950 - val_loss: 0.0012 - val_acc: 0.7850
Epoch 61/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8041Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 62/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8017Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8043 - val_loss: 0.0012 - val_acc: 0.7710
Epoch 63/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7915Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7932 - val_loss: 0.0011 - val_acc: 0.7804
Epoch 64/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8017Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8014 - val_loss: 0.0012 - val_acc: 0.7827
Epoch 65/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7945Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7926 - val_loss: 0.0012 - val_acc: 0.7617
Epoch 66/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8083Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8061 - val_loss: 0.0012 - val_acc: 0.7874
Epoch 67/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7891Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7891 - val_loss: 0.0012 - val_acc: 0.7780
Epoch 68/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7963Epoch 00067: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7961 - val_loss: 0.0012 - val_acc: 0.7874
Epoch 69/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8179Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8178 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 70/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8041Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8049 - val_loss: 0.0012 - val_acc: 0.7757
Epoch 71/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8017Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8032 - val_loss: 0.0012 - val_acc: 0.7850
Epoch 72/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7975Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7996 - val_loss: 0.0012 - val_acc: 0.7827
Epoch 73/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7945Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7956 - val_loss: 0.0012 - val_acc: 0.7734
Epoch 74/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8005Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.7757
Epoch 75/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7981Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7973 - val_loss: 0.0011 - val_acc: 0.7874
Epoch 76/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7867Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7868 - val_loss: 0.0012 - val_acc: 0.7850
Epoch 77/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8071Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8078 - val_loss: 0.0011 - val_acc: 0.7780
Epoch 78/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8023Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8026 - val_loss: 0.0012 - val_acc: 0.7804
Epoch 79/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8023Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.8020 - val_loss: 0.0012 - val_acc: 0.7804
Epoch 80/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8017Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8020 - val_loss: 0.0012 - val_acc: 0.7921
Epoch 81/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8035Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7991 - val_loss: 0.0012 - val_acc: 0.7780
Epoch 82/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8101Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7850
Epoch 83/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8119Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8119 - val_loss: 0.0012 - val_acc: 0.7921
Epoch 84/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8065Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8072 - val_loss: 0.0012 - val_acc: 0.7944
Epoch 85/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7981Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7996 - val_loss: 0.0012 - val_acc: 0.7874
Epoch 86/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7963Epoch 00085: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7973 - val_loss: 0.0011 - val_acc: 0.7967
Epoch 87/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8053Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8067 - val_loss: 0.0011 - val_acc: 0.7664
Epoch 88/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8059Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8078 - val_loss: 0.0012 - val_acc: 0.7991
Epoch 89/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7981Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8008 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 90/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8029Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8014 - val_loss: 0.0011 - val_acc: 0.7804
Epoch 91/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8005Epoch 00090: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.7991 - val_loss: 0.0012 - val_acc: 0.7897
Epoch 92/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7981Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.7985 - val_loss: 0.0012 - val_acc: 0.7874
Epoch 93/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8035Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8055 - val_loss: 0.0011 - val_acc: 0.7850
Epoch 94/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8185Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0011 - acc: 0.8183 - val_loss: 0.0012 - val_acc: 0.8061
Epoch 95/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8095Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 3s - loss: 0.0011 - acc: 0.8078 - val_loss: 0.0012 - val_acc: 0.7897
Epoch 96/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8023Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8037 - val_loss: 0.0012 - val_acc: 0.7850
Epoch 97/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8185Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8201 - val_loss: 0.0011 - val_acc: 0.7991
Epoch 98/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8083Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8084 - val_loss: 0.0011 - val_acc: 0.7827
Epoch 99/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7993Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8026 - val_loss: 0.0011 - val_acc: 0.7897
Epoch 100/100
1664/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8119Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 4s - loss: 0.0010 - acc: 0.8119 - val_loss: 0.0012 - val_acc: 0.7944

---

Step 7: Visualize the Loss and Test Predictions

(IMPLEMENTATION) Answer a few questions and visualize the loss

Question 1: Outline the steps you took to get to your final neural network architecture and your reasoning at each step.

Answer: I have started with an architecture similar to famous computer vision designs like VGG-Net. It had 5 or 6 layer which each one has a Convolutional Layer, followed by a Max-Pooling Layer, and a Dropout Layer after the Max-Pooling. After that, Flattened output is feed to 3 layer of Fully Connected Dense Layer for classification. I have reduced the model to 3 layer of Convolutional (Convolutional Layer, Max-Pooling and Dropout) and 3 layer of Dense Layer after overfitting detected.

Question 2: Defend your choice of optimizer. Which optimizers did you test, and how did you determine which worked best?

Answer: My model has been run with all optimizers provided by Keras. I selected the optimizer with provide higher accuracy and lower loss and converge sooner. RMSprop converges after around 20 epochs and provide accuracy close to 80%. So I chose RMSprop as my model optimizer.

Use the code cell below to plot the training and validation loss of your neural network. You may find this resource useful.

## TODO: Visualize the training and validation loss of your neural network
for opt in opts:
    h = hists[opt]    
    plt.plot(h.history['loss'])
    plt.plot(h.history['val_loss'])
    plt.title('model loss ' + opt)
    plt.ylabel('Loss')
    plt.xlabel('epoch')
    plt.legend(['train', 'test'], loc='upper left')
    plt.show()
    plt.plot(hists[opt].history['acc'])
    plt.plot(hists[opt].history['val_acc'])
    plt.title('model accuracy ' + opt)
    plt.ylabel('Accuracy')
    plt.xlabel('epoch')
    plt.legend(['train', 'test'], loc='upper left')
    plt.show()
    
for opt in opts:
    h = hists[opt]
    plt.plot(h.history['val_loss'])
plt.title('models loss')
plt.ylabel('Loss')
plt.xlabel('epoch')
plt.legend(opts, loc='upper left')
plt.show()

for opt in opts:
    h = hists[opt]
    plt.plot(h.history['val_acc'])

plt.title('models accuracy')
plt.ylabel('Accuracy')
plt.xlabel('epoch')
plt.legend(opts, loc='upper left')
plt.show()

Question 3: Do you notice any evidence of overfitting or underfitting in the above plot? If so, what steps have you taken to improve your model? Note that slight overfitting or underfitting will not hurt your chances of a successful submission, as long as you have attempted some solutions towards improving your model (such as regularization, dropout, increased/decreased number of layers, etc).

Answer: Yes, I faced some overfitting and I notice that my model is too complex. I reduce the number of layer in my model.

# Select the final model
final_model = "RMSprop"
model.load_weights(final_model + "_weights_best.hdf5")

Visualize a Subset of the Test Predictions

Execute the code cell below to visualize your model's predicted keypoints on a subset of the testing images.

y_test = model.predict(X_test)
fig = plt.figure(figsize=(20,20))
fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)
for i in range(9):
    ax = fig.add_subplot(3, 3, i + 1, xticks=[], yticks=[])
    plot_data(X_test[i], y_test[i], ax)

---

Step 8: Complete the pipeline

With the work you did in Sections 1 and 2 of this notebook, along with your freshly trained facial keypoint detector, you can now complete the full pipeline. That is given a color image containing a person or persons you can now

• Detect the faces in this image automatically using OpenCV
• Predict the facial keypoints in each face detected in the image
• Paint predicted keypoints on each face detected

In this Subsection you will do just this!

(IMPLEMENTATION) Facial Keypoints Detector

Use the OpenCV face detection functionality you built in previous Sections to expand the functionality of your keypoints detector to color images with arbitrary size. Your function should perform the following steps

1. Accept a color image.
2. Convert the image to grayscale.
3. Detect and crop the face contained in the image.
4. Locate the facial keypoints in the cropped image.
5. Overlay the facial keypoints in the original (color, uncropped) image.

Note: step 4 can be the trickiest because remember your convolutional network is only trained to detect facial keypoints in $96 \times 96$ grayscale images where each pixel was normalized to lie in the interval $[0,1]$, and remember that each facial keypoint was normalized during training to the interval $[-1,1]$. This means - practically speaking - to paint detected keypoints onto a test face you need to perform this same pre-processing to your candidate face - that is after detecting it you should resize it to $96 \times 96$ and normalize its values before feeding it into your facial keypoint detector. To be shown correctly on the original image the output keypoints from your detector then need to be shifted and re-normalized from the interval $[-1,1]$ to the width and height of your detected face.

When complete you should be able to produce example images like the one below

# Load in color image for face detection
image = cv2.imread('images/obamas4.jpg')


# Convert the image to RGB colorspace
image_copy = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)


# plot our image
fig = plt.figure(figsize = (9,9))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('image copy')
ax1.imshow(image_copy)

<matplotlib.image.AxesImage at 0x7f869d42ff28>

### TODO: Use the face detection code we saw in Section 1 with your trained conv-net 
## TODO : Paint the predicted keypoints on the test image
def key_facial_points(color_image):
    color_image = cv2.cvtColor(color_image, cv2.COLOR_BGR2RGB)
    image_with_key_points = np.copy(color_image)
    
    # Convert the image to grayscale.
    gray = cv2.cvtColor(color_image, cv2.COLOR_RGB2GRAY)
    
    # Detect and crop the face contained in the image.
    # Extract the pre-trained face detector from an xml file
    face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')
    # Detect the faces in image
    faces = face_cascade.detectMultiScale(gray)
    
    all_key_point = []
    # crop faces in image
    for (x, y, w, h) in faces:
        scale = w / 96.0
        
        f = gray[y:y+h, x:x+w]
        resized_f = cv2.resize(f,(96,96)) # resize to 96*96
        resized_f = resized_f / 255.0 # normalized to [0,1]
        resized_f=resized_f.reshape(1,96,96,1)
        
        # Locate the facial keypoints in the cropped image.
        # detect the key points
        prediction = model.predict(resized_f)

        prediction = prediction * 48 + 48 # undo the normalization
        prediction = prediction[0]
        
              
        
        # Overlay the facial keypoints in the original (color, uncropped) image.
        # mark the key points in image_with_key_points
        for i in range(0,len(prediction),2):
            x_k = int(prediction[i] * scale) + x
            y_k = int(prediction[i+1] * scale) + y
            all_key_point.append((x_k,y_k))
            
      
        
        
    # Overlay the facial keypoints in the original (color, uncropped) image.
    fig = plt.figure(figsize = (9,9))
    ax1 = fig.add_subplot(111)
    ax1.set_xticks([])
    ax1.set_yticks([])
    ax1.set_title('Image with face key points')
    # draw the rectangle around the faces
    for (x, y, w, h) in faces:
        cv2.rectangle(image_with_key_points, (x,y), (x+w,y+h), (255,0,0), 3)
    # specify the key points on all faces
    for key_point in all_key_point:
        cv2.circle(image_with_key_points, key_point, 1, (0, 255, 0), 2)
    ax1.imshow(image_with_key_points)

key_facial_points(image)

(Optional) Further Directions - add a filter using facial keypoints to your laptop camera

Now you can add facial keypoint detection to your laptop camera - as illustrated in the gif below.

The next Python cell contains the basic laptop video camera function used in the previous optional video exercises. Combine it with the functionality you developed for keypoint detection and marking in the previous exercise and you should be good to go!

import cv2
import time 
from keras.models import load_model
def laptop_camera_go():
    # Create instance of video capturer
    cv2.namedWindow("face detection activated")
    vc = cv2.VideoCapture(0)

    # Try to get the first frame
    if vc.isOpened(): 
        rval, frame = vc.read()
    else:
        rval = False
    
    # keep video stream open
    while rval:
        # plot image from camera with detections marked
        cv2.imshow("face detection activated", frame)
        
        # exit functionality - press any key to exit laptop video
        key = cv2.waitKey(20)
        if key > 0: # exit by pressing any key
            # destroy windows
            cv2.destroyAllWindows()
            
            # hack from stack overflow for making sure window closes on osx --> https://stackoverflow.com/questions/6116564/destroywindow-does-not-close-window-on-mac-using-python-and-opencv
            for i in range (1,5):
                cv2.waitKey(1)
            return
        
        # read next frame
        time.sleep(0.05)             # control framerate for computation - default 20 frames per sec
        rval, frame = vc.read()  

# Run your keypoint face painter
laptop_camera_go()

(Optional) Further Directions - add a filter using facial keypoints

Using your freshly minted facial keypoint detector pipeline you can now do things like add fun filters to a person's face automatically. In this optional exercise you can play around with adding sunglasses automatically to each individual's face in an image as shown in a demonstration image below.

To produce this effect an image of a pair of sunglasses shown in the Python cell below.

# Load in sunglasses image - note the usage of the special option
# cv2.IMREAD_UNCHANGED, this option is used because the sunglasses 
# image has a 4th channel that allows us to control how transparent each pixel in the image is
sunglasses = cv2.imread("images/sunglasses_4.png", cv2.IMREAD_UNCHANGED)

# Plot the image
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.imshow(sunglasses)
ax1.axis('off');

This image is placed over each individual's face using the detected eye points to determine the location of the sunglasses, and eyebrow points to determine the size that the sunglasses should be for each person (one could also use the nose point to determine this).

Notice that this image actually has 4 channels, not just 3.

# Print out the shape of the sunglasses image
print ('The sunglasses image has shape: ' + str(np.shape(sunglasses)))

It has the usual red, blue, and green channels any color image has, with the 4th channel representing the transparency level of each pixel in the image. Here's how the transparency channel works: the lower the value, the more transparent the pixel will become. The lower bound (completely transparent) is zero here, so any pixels set to 0 will not be seen.

This is how we can place this image of sunglasses on someone's face and still see the area around of their face where the sunglasses lie - because these pixels in the sunglasses image have been made completely transparent.

Lets check out the alpha channel of our sunglasses image in the next Python cell. Note because many of the pixels near the boundary are transparent we'll need to explicitly print out non-zero values if we want to see them.

# Print out the sunglasses transparency (alpha) channel
alpha_channel = sunglasses[:,:,3]
print ('the alpha channel here looks like')
print (alpha_channel)

# Just to double check that there are indeed non-zero values
# Let's find and print out every value greater than zero
values = np.where(alpha_channel != 0)
print ('\n the non-zero values of the alpha channel look like')
print (values)

This means that when we place this sunglasses image on top of another image, we can use the transparency channel as a filter to tell us which pixels to overlay on a new image (only the non-transparent ones with values greater than zero).

One last thing: it's helpful to understand which keypoint belongs to the eyes, mouth, etc. So, in the image below, we also display the index of each facial keypoint directly on the image so that you can tell which keypoints are for the eyes, eyebrows, etc.

With this information, you're well on your way to completing this filtering task! See if you can place the sunglasses automatically on the individuals in the image loaded in / shown in the next Python cell.

# Load in color image for face detection
image = cv2.imread('images/obamas4.jpg')

# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)


# Plot the image
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Original Image')
ax1.imshow(image)

## (Optional) TODO: Use the face detection code we saw in Section 1 with your trained conv-net to put
## sunglasses on the individuals in our test image

(Optional) Further Directions - add a filter using facial keypoints to your laptop camera

Now you can add the sunglasses filter to your laptop camera - as illustrated in the gif below.

The next Python cell contains the basic laptop video camera function used in the previous optional video exercises. Combine it with the functionality you developed for adding sunglasses to someone's face in the previous optional exercise and you should be good to go!

import cv2
import time 
from keras.models import load_model
import numpy as np

def laptop_camera_go():
    # Create instance of video capturer
    cv2.namedWindow("face detection activated")
    vc = cv2.VideoCapture(0)

    # try to get the first frame
    if vc.isOpened(): 
        rval, frame = vc.read()
    else:
        rval = False
    
    # Keep video stream open
    while rval:
        # Plot image from camera with detections marked
        cv2.imshow("face detection activated", frame)
        
        # Exit functionality - press any key to exit laptop video
        key = cv2.waitKey(20)
        if key > 0: # exit by pressing any key
            # Destroy windows 
            cv2.destroyAllWindows()
            
            for i in range (1,5):
                cv2.waitKey(1)
            return
        
        # Read next frame
        time.sleep(0.05)             # control framerate for computation - default 20 frames per sec
        rval, frame = vc.read()    
        

# Load facial landmark detector model
model = load_model('my_model.h5')

# Run sunglasses painter
laptop_camera_go()